• Colloquium Summer Term 2023

    The CeNoS Colloquium starts at 4.30 p.m. as zoom session. If you are interested in the zoom link, please send an email to okamp@uni-muenster.de


    Applications of AI




  • Colloquium Winter Term 2022/2023

    The CeNoS Colloquium starts at 4.30 p.m. as zoom session. If you are interested in the zoom link, please send an email to okamp@uni-muenster.de


    Online Colloquium / ZOOM Session

    Collocation-Based EEG/MEG Inverse Problem Solution Based on Helmholtz Reciprocity Principle: The Same MNE?

    Sergey N Makarov
    Worcester Polytechnic Institute (Worcester MA) and Martinos Faculty at Massachusetts General Hospital (Charlestown MA)

    Helmholtz reciprocity principle makes it possible to expand an unknown cortical dipole density into global bases - cortical fields of sequentially excited electrodes (EEG) or coils (MEG). Some results will be given and a connection to the minimum norm estimation (MNE) formulation will be discussed.


    Online Colloquium / ZOOM Session

    Critical Transitions in the Earth System

    Niklas Boers
    Professur für Earth System Modelling, TUM School of Engineering and Design, Technische Universität München

    In response to anthropogenic release of greenhouse gases, the Earth is warming at unprecedented rates. It has been suggested that several components of the Earth’s climate system may respond with abrupt transitions between alternative stable states in response to gradual changes in forcing. Based on the theory of stochastically forced dynamical systems and their bifurcations, a methodology is presented to measure changes in the stability of a given equilibrium state from observational time series. The method is applied to observation-based data to investigate if and how the stability of the Greenland Ice Sheets, the Atlantic Meridional Overturning Circulation, the Amazon rainforest, and the South American monsoon, has changed in the course of the last century.


    Online Colloquium / ZOOM Session

    The German Climate Computing Center - In-situ Data Processing of Large Climate Model Data

    Nils-Arne Dreier
    Deutsches Klimarechenzentrum, Hamburg

    The German Climate Computing Center (DKRZ) provides the central research infrastructure for the German climate and earth system research community. Besides operating the supercomputer, DKRZ also provides support for users and contributes to the development of climate models. Based on a brief introduction to the climate model ICON, I will present my work at DKRZ as a research software engineer, which concerns the in-situ visualization and analysis of large climate model data. With higher resolutions the amount of data that is produced by the simulation increases. This makes traditional data processing workflows unusable, as they include the storage on a hard drive. We are following two strategies, the first approach develops an interconnection to paraview via the catalyst interface. This enables us to produce high-quality 3d visualizations of the high-resolution data. In the second approach, we develop a python interface for domain scientists, enabling them to execute their analyses and visualization in an in-situ fashion.


    Interfaces and Dynamical Transitions in Strongly Interacting Many-body Systems

    Aljaz Godec
    Mathematical bioPhysics Group, Max Planck Institute for Multidisciplinary Sciences, Göttingen

    Our understanding of collective phenomena, especially in inhomogeneous systems, is mostly based on mean-field ideas that neglect correlations, whose importance in turn grows with the strength of interactions. This raises doubts whether mean-field ideas sensibly describe the physical behavior of strongly interacting systems. In the talk I will describe our recent efforts to account for and understand pair-correlation effects in classical many-body systems in and out of equilibrium. I will illuminate unexpected effects of correlations in the kinetics [1] and interfacial phenomena [2], and show that they give rise to global speed limits during non-equilibrium relaxation [3].

    [1] K. Blom & AG, Phys. Rev. X 11, 031067 (2021)

    [2] K. Blom, N. Ziethen, D. Zwicker, & AG, arXiv:2204.02962 (2022).

    [3] K. Blom & AG, arXiv:2209.14287 (2022)


    Nuclei Coordinate Cell Division in Space and Time

    Lendert Gelens, PhD
    Laboratory of Dynamics in Biological Systems, KU Leuven

    Upon fertilization, the early Xenopus laevis frog egg quickly divides about ten times to go from a single cell with a diameter of a millimeter to several thousands of cells of somatic cell size. Using cytoplasmic extracts made from frog eggs, many cellular processes have been reconstituted and studied in vitro. Homogenized extracts can also spontaneously self-organize into various cellular spatial structures when mixing isolated components back together. 

    Here, we will discuss our recent experimental and modeling efforts to understand how biochemical oscillations and waves form in the frog cytoplasm, and how they help organize the cell division cycle. We study an artificial cell system consisting of droplets of frog egg extracts of varying sizes, which allows characterizing the spatiotemporal dynamics of cell cycle oscillations, periodic nuclear assembly and destruction, and cytoskeletal organization. We complement our experimental analysis by developing computational models incorporating nuclear dynamics, illustrating the role of the nucleus in setting the cell cycle oscillation period. Nuclei are shown to function as pacemakers, regions which oscillate faster than its surroundings. As a result, nuclei send out biochemical waves that can synchronize the whole medium and ensure that cell division is properly coordinated in the large frog egg. 


    Quasi-Stationarities Relative to the Collective Market Motion in Financial Markets

    Anton Heckens
    Universität Duisburg-Essen, Fakultät für Physik

    Financial markets are non-stationary and their dynamics is dominated by strong collective effects. When studying correlation matrices in subsequent epochs, quasi-stationary patterns,
    referred to as market states, are detected by applying clustering methods. They emerge, disappear or reemerge, but they are dominated by the collective motion of all stocks. By removing the dominating market motion in a proper way, we uncover different market states
    which are stable over a long period of time.
    We discuss their precursor properties in the US stock markets and collective effects relative to the dominating collective market behavior.


  • Colloquium Summer Term 2022

    The CeNoS Colloquium starts at 16.30 s.t as zoom session. If you are interested in the zoom link, please send an email to okamp@uni-muenster.de


    Online Colloquium / ZOOM Session

    Data-Driven Analysis of Complex Systems for a Sustainable Future

    Dr. Benjamin Schäfer
    Karlsruher Institut für Technologie (KIT) / Institut für Automation und angewandte Informatik (IAI)

    The transition to a sustainable energy system raises numerous complex challenges, as power generation becomes more uncertain, while simultaneously more operational data becomes available. Still, a coherent approach to fully utilize such data is missing. Within this talk, I will sketch our efforts in developing a data-driven framework, combining data analysis, machine learning and modelling approaches to study complex (energy) systems. I will emphasize the need for open data but also for transparency in physical and machine (learning) models. In particular, I will show how we may extract scaling laws, obtain forecasts, and identify drivers in different power grids from a purely data-driven perspective.


    Online Colloquium / ZOOM Session

    Modelling and controlling turbulent vortex dynamics in cardiac systems and active fluids

    Prof. Dr. Markus Bär
    Physikalisch-Technische Bundesanstalt (PTB), Berlin 

    Many relevant experimental systems and applications exhibit irregular dynamics such as spatiotemporal chaos or turbulence.  In this talk, I will present two examples of such behavior where control of such states by transforming their dynamics into homogeneous or periodic steady states or regular periodic patterns is desirable. The specific applications are (i) defibrillation of cardiac tissue by periodic pacing and (ii) turbulence in active suspensions. For the cardiac system, we used simulations as well as data analysis not only to identify suitable conditions e. g. pacing strength and periods [1, 2], but also to determine the mechanism of defibrillation in this specific modus.  For active turbulence, a simple phenomenological deterministic continuum model was used for quantitative modelling of experimental control in bacterial suspensions. Detailed investigations showed how the turbulent collective dynamics could be transformed into more regular, periodic spatiotemporal patterns of different symmetries using a periodic array of obstacles. This is in line with recent experimental findings using confinement by an arrays of pillars [3].  A recent study addresses the competition between the controlling effect of confinement and a nonlinear advection term behind the emergence of turbulence in active fluids demonstrating that the breakdown of control occurs through an Ising-like phase transition in a coarse-grained description of the model [4].



    [1] P. Buran, M. Bär, S. Alonso and T. Niedermayer. Chaos 27, 113110 (2017).

    [2] P. Buran, T Niedermayer, and M Bär. Preprint on https://www.biorxiv.org (2021).

    [3] H. Reinken, D. Nishiguchi, S. Heidenreich, A. Sokolov, M. Bär, S. H. L. Klapp, and I. Aranson. Comm. Phys. 3, 76 (2020).

    [4] H. Reinken, S. Heidenreich, M. Bär, S. Klapp. Phys. Rev. Lett. 128, 048004 (2022).



    Online Colloquium / ZOOM Session

    Machine Learning for Nanophotonics and Nanophotonics for Machine Learning

    Prof. Dr. Carsten Schuck
    Physikalisches Institut  (WWU) /  Center for NanoTechnology (CeNTech) / Center for Soft Nanoscience (SoN)

    Machine Learning techniques offer novel approaches to a wide range of complex problems that may defy human intuition. One area where this is particularly relevant is the field of nanophotonic design. Nanostructured Photonic Integrated Circuits (PICs) are replacing their electrical equivalents in high bandwidth telecommunication as well as sensing applications and play a key role in modern quantum technology. However, designing nanophotonic circuit components today relies primarily on combining intuitive concepts, such as interference and coupled mode theory, with brute force parameter optimization. In view of the vastness of possible device configurations, such conventional design approaches only consider an infinitesimal part of the solution space. We here show how to go beyond this paradigm by employing an inverse design technique that relies on reinforcement learning for finding non-intuitive configurations of nanophotonic devices that exceed conventional designs in both performance and compactness.

    Nanophotonics, on the other hand, also provides means for novel hardware implementations of neural networks. Optical artificial neural networks (OANNs) allow for extremely energy-efficient signal processing at the speed of light, therewith overcoming the performance limitations of corresponding implementations on electronic hardware, such as CPUs or GPUs, that are subject to the von Neumann bottleneck. Here we show a path towards realizing the linear and nonlinear building blocks of OANNs on a nanophotonic platform that could be scaled to large system size, thus realizing a photonic neural network processing unit. In particular, we leverage a variety of molecular compounds developed in the context of the collaborative research center on “Intelligent Matter” (CRC 1459) to realize nonlinear activation functions and develop interfaces between nanophotonic OANN components and photoresponsive molecular systems. We further study the performance of such nanophotonic neural networks that are trained in the presence of noise, naturally occurring in ensembles of molecular compounds but absent in digital electronic implementations.


    Online Colloquium / ZOOM Session

    What is the Use of Nonlinear Dynamics in Practice? - Use Cases from Critical Infrastructures and Life Sciences

    Dr. Arndt Telschow
    Cuculus GmbH, Erfurt

    Nonlinear dynamics and nonlinear effects are ubiquitous in complex systems, including ecosystems, technical and social systems. In the lecture, I will use selected examples from critical infrastructure and the life sciences to show the great benefits that nonlinear approaches can have in practice. The use cases are partly based on my work at the Westfälische Wilhelms-Universität Münster and partly on my current work at the software company Cuculus (www.cuculus.com). Use cases discussed include: (1) Forecasting. I present the ZONOS 24insight software module for forecasting electricity consumption and generation (see https://24insight.zonos.de for a live demo). (2) Fraud Detection. I present the results of a white paper describing a new approach to identifying electricity theft. (3) Anticipate critical transitions. I present a new method for quantifying resilience in complex systems and show how it can be used to predict tipping points. By analyzing empirical data, the method is used to anticipate (i) power grid collapse, (ii) global climate change (end of last greenhouse earth), bacterial population collapse, physiological changes and intracellular switching.


    Online Colloquium / ZOOM Session

    Design and Theory of Self-Avoiding Active Matter

    Dr. Mazi Jalaal
    University of Amsterdam 

    Active matters present systems made of micro-scale agents that consume energy from the environment. In many types of such systems, the active agent leaves a trace behind, creating memory in space and time. In this talk, I will first give a brief overview of different classes of active matter with such a feature and then present a frugal and tuneable experimental framework in which one could synthetically design active particles with a self-avoiding memory. I will then present a theoretical framework based on mathematical billiards and show how active matter with memory can result in complex, chaotic patterns in a simple deterministic dynamical system. 


    Additional date / Two talks

    1. Talk:  On-off Intermittency due to Parametric Lévy Noise

    Dr. Adrian van Kan
    University of California, Berkeley | UCB · Department of Physics 

    Instabilities arise in many physical systems at some parameter threshold. Typically the system is embedded in an uncontrolled noisy environment. The fluctuating properties of the environment affect the control parameters of the instability, which leads to multiplicative noise. The result of multiplicative noise close to an instability threshold is on-off intermittency, which is characterised by an aperiodic switching between a large-amplitude “on” state and a small-amplitude “off” state.
    I present a new type of intermittency, Lévy on-off intermittency, which arises from multiplicative α-stable white noise close to an instability threshold [1]. We study this problem in the linear and nonlinear regimes, both theoretically and numerically, for the case of a pitchfork bifurcation with fluctuating growth rate. In a recently introduced point-vortex model of 3-D perturbations in 2-D flows [2], the perturbation amplitude obeyed  such an equation. We compute the stationary distribution analytically and numerically from the associated fractional Fokker-Planck equation in the Stratonovich interpretation.
    We characterize the system in the parameter space (α, β) of the noise, with stability parameter α ∈ (0, 2) and skewness parameter β ∈ [−1, 1]. Five regimes are identified in this parameter space, in addition to the well-studied Gaussian case α = 2. Three regimes are located at 1 < α < 2, where the noise has finite mean but infinite variance. They are differentiated by β and all display a critical transition at the deterministic instability threshold, with on-off intermittency close to onset. Critical exponents are computed from the stationary distribution. Each regime is characterised by a specific form of the density and specific critical exponents, which differ starkly from the Gaussian case. A finite or infinite number of moments may converge, depending on parameters. Two more regimes are found at 0 < α ≤ 1. There, the mean of the noise diverges, and no transition occurs. In one case the origin is always unstable, independently of the distance μ from the deterministic threshold. In the other case, the origin is conversely always stable, independently of μ. In summary, we demonstrate that an instability subject to non-equilibrium, power-law-distributed fluctuations can display substantially different properties than for Gaussian thermal fluctuations, in terms of statistics and critical behavior.

    [1]  van Kan, A., Alexakis, A. and Brachet, M.E., 2021. Lévy on-off intermittency. Physical Review E, 103(5), p.052115.
    [2] van Kan, A., Alexakis, A. and Brachet, M.E., 2021. Intermittency of three-dimensional perturbations in a point-vortex model. Physical Review E, 103(5), p.053102.


    2. Talk: Pressure-driven Wrinkling of Soft Inner-lined Tubes

    Dr. Ben Foster
    University of California, Berkeley | UCB · Department of Physics 

    A simple equation modelling an inextensible elastic lining of an inner-lined tube subject to an imposed pressure difference is derived from a consideration of the idealised elastic properties of the lining and the pressure and soft-substrate forces. Two cases are considered in detail, one with prominent wrinkling and a second one in which wrinkling is absent and only buckling remains. Bifurcation diagrams are computed via numerical continuation for both cases. Wrinkling, buckling, folding, and mixed-mode solutions are found and organised according to system-response measures including tension, in-plane compression, maximum curvature and energy. Approximate wrinkle solutions are constructed using weakly nonlinear theory, in excellent agreement with numerics. Our approach explains how the wavelength of the wrinkles is selected as a function of the parameters in compressed wrinkling systems and shows how localised folds and mixed-mode states form in secondary bifurcations from wrinkled states. Our model aims to capture the wrinkling response of arterial endothelium to blood pressure changes but applies much more broadly.



  • Colloquium Winter Term 2021/2022

    The CeNoS Kolloquium starts at 16.30 s.t as zoom session. If you are interested in the zoom link, please send an email to okamp@uni-muenster.de


    Smart grid communication and cyber security: a computer-science perspective

    Prof. Anna Remke
    Institut für Informatik, WWU

    Within smart grids the safe and dependable distribution of electric power highly depends on the security of Supervisory Control and Data Acquisition (SCADA) systems and their underlying communication protocols. Existing network-based intrusion detection systems for Industrial Control Systems (ICS) are usually centrally applied at the SCADA server and do not take the underlying physical process into account. A recent line of work proposes an additional layer of security via a process-aware approach applied locally at the field stations. We broaden the scope of process-aware monitoring by considering the interaction between neighboring field stations, which facilitates upcoming trends of decentralized energy management (DEM). Local security monitoring is lifted to monitoring neighborhoods of field stations, therefore achieving a broader grid coverage w.r.t. security.


    The role of non-conservative interactions in nonequilibrium complex systems

    Dr. Sarah Loos
    International Center for Theoretical Physics, Trieste, Italy

    The complex world surrounding us, including all living matter and various artificial complex systems, is mostly far from thermal equilibrium. A major goal of modern statistical physics and thermodynamics is to unravel the fundamental principles and in particular the collective phenomena of such nonequilibrium systems, like the swarming of fish, flocking of birds, or pedestrian crowd dynamics. A key novel concept to describe and classify nonequilibrium systems is the stochastic entropy production, which explicitly quantifies the breaking of time-reversal symmetry. However, so far, little attention has been paid to the implications of non-conservative interactions, such as time-delayed (i.e., retarded) or nonreciprocal interactions, which cannot be represented by Hamiltonians contrasting all interactions traditionally considered in statistical physics. Time-delayed and nonreciprocal interactions indeed emerge commonly in biological, chemical and feedback systems, and are widespread in engineering and machine learning. In this talk, I will use simple time- and space-continuous models to discuss technical challenges and unexpected physical phenomena induced by nonreciprocity [1,2] and time delay [3,4].

    [1] Loos and Klapp, NJP 22, 123051 (2020)
    [2] Loos, Hermann, and Klapp, Entropy 23, 696 (2021)
    [3] Loos and Klapp, Sci. Rep. 9, 2491 (2019)
    [4] Holubec, Geiss, Loos, Kroy, and Cichos, Phys. Rev. Lett. in press (2021)


    Complexity meets energy: A slightly biased, high-variance survey on modeling power grid dynamics

    Dr. Katrin Schmietendorf
    Center for Nonlinear Science, WWU

    In the course of the energy transition the ‚traditional‘ power grid with large conventional plants injecting constant feed-in adjustable to the current demand is progressively being transformed into a decentralized grid with small and medium generating units, many of which are volatile and display feed-in variations on time scales ranging from annual to sub-seconds. This development poses novel challenges with respect to stable and reliable grid operation, frequency quality, storage facilities, control schemes and grid design. Against this backdrop, power system analysis has evolved from a primarily engineering topic to an interdisciplinary research area involving, inter alia, statistics, meteorology, computer science, turbulence research, network and complex systems theory.

    About a decade ago, Filatrella et al. uncovered that the most basic generator model standardly used in electrical engineering corresponds to a modification of the Kuramoto model, a paradigmatic model for self-organized synchronization. Building on this relationship, a new branch of research, also referred to as Kuramoto-like power grid modeling, evolved within the nonlinear dynamics community. In Kuramoto-like power grid modeling power systems are analyzed as networks of coupled oscillators. Various application-oriented issues have been addressed so far, e. g. power outages due to severe perturbations and transmission-line overloads, cascading failures, the impact of fluctuating feed-in on power quality and grid stability, storage and new control strategies.

    The purpose of my talk is to give a survey across the major branches of Kuramoto-like power grid modeling, summarize the main results obtained so far and attempt an outlook to future focuses of research, both thematically and methodologically.  Despite its summary character, the talk will put emphasis on findings me and collaborators obtained during my doctoral research which concern the impact of short-term wind power fluctuations on grid stability and power quality. The overview and outlook is also intended to serve as basis for identifying potential synergy effects and points of contact for follow-up research projects at the WWU.


  • Colloquium Summer Term 2021


    Social insects as models for complex systems?

    Prof. Jürgen Gadau
    Institute for Evolution and Biodiversity , WWU

    Social complexity is a central theme in complex system theory, biological sciences, and social sciences, and yet features that underlie it, and the evolutionary processes that shaped them, are not well understood. It is unclear what composition of features is needed for the emergence of a high-level social organization, and what are their relative weights. The broad diversity of social structures in insects (such as bees, ants, termites) presents a unique opportunity to study the features that underlie social complexity, and their evolutionary trajectories.


    Artificial Intelligence: Theory or Tool? A Brief Discussion Drawn From Interdisciplinary Machine Learning Research Examples.

    Jun. Prof. Benjamin Risse
    Institut für Informatik, WWU

    Artificial Intelligence (AI) is arguably the most discussed technology of our times: On the one hand, AI algorithms have led to a variety of substantial breakthroughs which have inspired the use of AI for many qualitatively different research questions. On the other hand, no other research paradigm has a comparable history in falling behind its theoretical considerations causing recurring 'AI winters'. In order to provide a balanced discussion I will present several successful deep learning applications which will naturally lead to intrinsic challenges and limitations of state-of-the-art AI algorithms. These limitations are particularly apparent in interdisciplinary projects in which AI is predominantly used as a tool to 'solve' a particular problem from the respective field. An in-depth investigating of the mathematical foundation of these algorithms is therefore pivotal to bridge the gap between theory and practice in order to save AI from its self-induced hypes. Therefore, the overall goal of my talk is to present some hopefully inspiring AI examples while advocating for truly interdisciplinary research projects beyond the borders of traditional academic disciplines.


    Mathematical and statistical models for causal inference in public health decision-making

    Prof. Dr. med. André Karch
    Institute of Epidemiology and Social Medicine, WWU

    The current SARS-CoV-2 pandemic reveals the importance of evidence-based decision-making in public health. The underlying research questions are often complex, need to be assessed in real-time and need to rely on routinely collected data. Dependent on the research question, dynamic mathematical models, statistical models and modern epidemiological concepts need to be applied to come into a position where causal inference is possible. In this talk I will present examples from our work from the fields of infectious diseases modelling, microbiome research and the evaluation of cancer screening programs to discuss challenges of and solutions for evidence-based decision-making in public health.


    Self-assembly in soft matter with multiple length scales

    Alberto Scacchi
    Department of Chemistry and Materials Science, Aalto University, Finnland

    Spontaneous self-assembly in molecular systems is a fundamental route to both biological and engineered soft matter. Simple micellization, emulsion formation, and polymer mixing are well understood. However, the principles behind emergence of structures with competing length scales in soft matter systems remain unknown. Examples include droplet-inside-droplet assembly in many biomacromolecular systems undergoing liquid-liquid phase separation, analogous multiple emulsion formation in oil-surfactant-water formulations, and polymer core-shell particles with internal structure.

    We developed a microscopic theoretical model based on effective interactions between the constituents of a soft matter system to explain self-organization both at single and multiple length scales.

    The findings provide guidelines to understanding the length scales rising spontaneously in biological self-assembly, but also open new venues to the development and engineering of biomolecular and polymeric functional materials and pharmaceutical formulations.


    Computational molecular design of bioactive compounds

    PD. Dr. Oliver Koch
    Institute of Pharmaceutical and Medicinal Chemistry, WWU

    My research interest lies in the development and application of computational methods in rational drug design with focus on structure-based design, ‘big data’ driven decisions and molecular machine learning in order to develop bioactive molecules and to understand selectivity and promiscuity of protein-ligand interactions. This work is generally aimed at applying and improving the success of computational methods in delivering novel and safe small molecule therapeutics


    Dynamic behavior of growth processes: Phase separation, self-similarity, and oscillations

    Prof. Dr. André Schlichting
    Arbeitsgruppe Dynamik komplexer Systeme, Mathematik, WWU

    The talk reviews some growth processes describing the evolution of clusters consisting of atomic parts called monomers. The growth and shrinkage can only occur by adding and removing single monomers, which is specified through a rate kernel depending solely on the involvedclusters' size.

    First, I discuss the exchange-driven growth model, which was recently obtained as the mean-field limit of stochastic particle systems (zero-range process). This model's longtime behavior can be described entirely under a detailed balance condition on the kernel. Here, the total mass density, determined by the initial data, acts as an order parameter, in which the system shows a phase separation.

    Next, we consider the model for a family of product kernels, which do not satisfy a detailed balance condition. After a suitable rescaling to self-similar variables, the equation becomes a discrete Laplace with a power-law as diffusion coefficient, which in particular degenerates at
    the origin. We will see that the solution converges to a stretched exponential self-similar profile.

    Lastly, we consider the now-classic Becker-Döring system to which an injection of monomers and a depletion of large clusters is added. These equations have been extensively used to model chemical-physical systems, especially bubbleator dynamics. By formal asymptotics, the model is approximated by a transport equation with a conservation law entering the boundary condition. For the limit model, a Hopf bifurcation is shown, indicating temporal oscillations in the model.


    Stretched Brownian Motion

    Prof. Dr. Martin Huesmann
    Institut für Mathematische Stochastik, WWU

    Given two marginal measures μ and ν we discuss in this talk how to find a natural martingale interpolation between μ and ν mimicking Brownian motion as closely as possible. Our approach relies on martingale versions of the Benamou-Brenier formula from optimal transport. A key step is the development of a dual theory which connects the optimal diffusion coefficients to solutions of the porous medium equation. In the end of the talk, we discuss how this interpolant can be viewed as a natural projection of Brownian motion onto the set of all martingales with initial law μ and terminal law ν wrt a metric that takes the information flow into account, the so called causal Wasserstein metric.


    Internal restructuring and reorganization in systems of soft and active matter

    Prof. Dr. Andreas Menzel
    Institut für Physik, Otto-von-Guericke-University Magdeburg

    The interplay between internal structure and internal or external driving can lead to pronounced nonlinear behavior in systems of soft or active matter. We concentrate on two examples of this kind.  First, we address the internal restructuring in magnetic gels or elastomers. These soft materials consist of magnetizable colloidal particles enclosed by an elastic polymer matrix. Driven by an external magnetic field and the induced magnetic interactions, the particles reposition against the elastic restoring forces of the surrounding medium. They can even reversibly collapse towards each other [1]. The process supports pronounced hysteresis [1], features interesting dynamics already in the overdamped regime and in simple one-dimensional settings [2], and affects the overall system behavior [3].
    Second, we investigate the collective dynamic behavior of self-propelled particles that migrate on a substrate and are assembled in regular, crystal-like arrangements [4]. Specifically, we focus on chiral active agents that feature a persistent bending in their trajectories, implying circular motion in the unperturbed case [5]. The circular motion of the individuals conflicts with the collective motion of the whole structure. With increasing chirality, collective motion thus breaks down. However, it can be restored by introducing phase boundaries into the system, leading, for example, to self-shearing or self-rotating crystallites.
    Future studies should, for instance, analyze in more detail the transitional regimes and the consequences for the overall dynamics in these situations.

    [1] M. Puljiz, S. Huang, K. A. Kalina, J. Nowak, S. Odenbach, M. Kästner, G. K. Auernhammer, A. M. Menzel, Soft Matter 14, 6809 (2018).
    [2] S. Goh, A. M. Menzel, H. Löwen, Phys. Chem. Chem. Phys. 20, 15037 (2018).
    [3] G. Pessot, M. Schümann, T. Gundermann, S. Odenbach, H. Löwen, A. M. Menzel, J. Phys.: Condens. Matter 30, 125101 (2018).
    [4] A. M. Menzel, T. Ohta, H. Löwen, Phys. Rev. E 89, 022301 (2014).
    [5] Z.-F. Huang, A. M. Menzel, H. Löwen, Phys. Rev. Lett. 125, 218002 (2020).


    Gibbs measures with space-time singularities

    Prof. Dr. Chiranjib Mukherjee
    Institut für Mathematische Stochastik, WWU

    Consider a finite volume Gibbs measure which descends from any translation-invariant interaction of Brownian paths in space and time. The class of interactions we are interested in include singularities in space and time (corresponding to the case of Fröhlich Polaron well-known in quantum mechanics) as well as a general class of Kac interactions arising in statistical mechanics. We show the existence of a unique Gibbs measure in the infinite volume (thermodynamic) limit and identify it explicitly. Next, in the strong coupling (resp. in the vanishing Kac) limit we show that the infinite volume limit coincides with the increments of the corresponding mean-field models. These results, which include various joint works with S.R.S. Varadhan (New York), confirm a series of conjectures of Spohn from 1980s.


    Stochastic dynamics in evolutionary ecology and gene regulation

    Dr. Peter Czuppon
    Institute for Evolution and Biodiversity, WWU

    The seminar gives a brief and non-exhaustive overview about non-linear dynamics in biological applications. I will focus on stochastic processes that describe dynamics of genetic variants (alleles) in a population of individuals of the same species. As an example, I will study evolutionary dynamics of (gametophytic) self-incompatibility (SI) alleles in a plant population. Self-incompatibility refers to the inability of self-reproduction, i.e., in order to produce plant offspring, the parental plant needs to be fertilized by a pollen with a different SI-allele. I study the (quasi-)stationary distribution of allele frequencies in a plant population of a fixed size and relate our findings to experimental data. As a second example of a non-linear stochastic process in biology, I will talk about auto-regulation in gene expression. Genes can self-regulate their expression, either by enhancing or reducing the transcription of a gene. These different regulation mechanisms lead to different gene expression profiles. A negative feedback (gene inhibition) reduces the variance of the number of produced proteins, while a positive feedback (genetic upregulation) strongly increases the variance. To quantify this effect precisely, we use an approximation based on the central limit theorem for Markov processes under a separation of time-scales.


  • Colloquium Winter Term 2020/2021


    Monitoring Subcellular Energy Physiology using Fluorescent Protein Biosensors

    Prof. Markus Schwarzländer
    Institut für Biologie und Biotechnologie der Pflanzen, WWU

    The energy conversion that occurs in cells requires tight surveillance and dynamic adjustment to meet demands, maintain efficiency and avoid dysfunction. This is particularly relevant in plant cells which are directly exposed to frequent and often dramatic changes in their immediate environment, including light-dark transitions or changes in oxygen availability. Yet, our understanding of the dynamics of energy physiology and their regulation at the sub-cellular or even sub-compartmental level is limited. We have been using quantitative confocal microscopy and fluorimetry to assess transitions in respiratory physiology in vivo using a growing set of genetically-encoded fluorescent protein sensors. I would like to introduce both fundamental considerations as well as the recent progress that we have made in the dissection of cellular energy physiology highlighting ATP dynamics, redox regulation and calcium transport. I will discuss our efforts towards multiparametric monitoring, as an approach towards an integrated picture of subcellular stress physiology, while appraising technical and biological limitations.


    The human dynamic system - Myonardo - a computational model of the human musculoskeletal system

    Prof. Heiko Wagner
    Institut für Sportwissenschaften, WWU

    In my lecture I will give a brief insight into the high complexity and non-linearity of the human musculoskeletal system. For the mathematical-physical description of human movement, we have developed a computer model that describes all important joints, segments and muscles in humans. With this computer model, the stress and strain within a person's body can now be analyzed. I will first introduce the Myonardo and then show a few examples of use, e.g. from the field of sports, ergonomics, and crash tests.


    PhD Kolloquium
    Machine Learning in der Medizin

    Lukas Fisch
    Medical Machine Learning Lab, WWU

    Age predicted from stochastic models using neuroimaging data of the brain has proven to be a useful biomarker to quantify the progress of neurological diseases and aging. However, most of these stochastic models rely on the neuroimaging data to be highly preprocessed, yielding a significant computational overhead and resulting in the predicted brain-age to be sensitive to preprocessing parameters. The few brain-age models which use so-called raw T1 neuroimaging data, rely on the voxels to be registered to a standard atlas which is still computationally expensive. In this work a 3D convolutional neural network (CNN) using the ResNet architecture is trained on raw non-registered T1 MRI neuroimaging data from N=10,691 samples from the German National Cohort and applied to N=4,004 samples from three independent studies. Additionally, seven models which are common in brain-age research are trained and validated on the same samples using preprocessed neuroimaging data. The 3D CNN predicts age with a median average deviation of as low as 2.59 years, outperforming all common brain-age models despite using raw non-registered brain scans.


    !!!! Postponed to 13.04.2021 !!!!

    Social insects as models for complex systems?

    Prof. Jürgen Gadau
    Institute for Evolution and Biodiversity , WWU

    Social complexity is a central theme in complex system theory, biological sciences, and social sciences, and yet features that underlie it, and the evolutionary processes that shaped them, are not well understood. It is unclear what composition of features is needed for the emergence of a high-level social organization, and what are their relative weights. The broad diversity of social structures in insects (such as bees, ants, termites) presents a unique opportunity to study the features that underlie social complexity, and their evolutionary trajectories.


  • Colloquium Winter Term 2018/2019

    23.10.2018 Mitgliederversammlung

    Neuronale Netze und Machine Learning - Einsatzgebiete in Industrie und Wirtschaft

    Uwe Westerhoff
    saracus consulting GmbH, Münster

    Das Themenfeld Machine Learning, Neuronale Netze und Künstliche Intelligenz hat insbesondere in den letzten kanpp zehn Jahren eine enorme Aufmerksamkeitssteigerung erfahren und dies sowohl im wissenschaftlichen Bereich als auch auf Seiten der Industrie. Inzwischen gibt es kaum eine Branche, welche nicht zumindest von sich behauptet, Verfahren der KI einzusetzen. Im Rahmen dieses Vortrags werden wir einen Blick hinter die Kulissen werfen.  Der erste Teil liefert einen Überblick über verschiedene Neuronale Netze und weitere Algorithmen. Außerdem werden wir uns zumindest kurz anschauen, wie Neuronale Netze letztlich auf Ebene der einzelnen Rechenoperationen arbeiten und trotz vergleichsweise einfacher Einzeloperationen ihre beeindruckende Leistungsfährigkeit erreichen.
    Im zweiten Teil stellen wir einige Anwendungsfälle aus Industrie und Wirtschaft vor und beschreiben die jeweils eingesetzten Verfahren.

    Einladender:  O. Kamps


    Bayesian Statistical Modeling for the Natural and Social Sciences

    Paul Buerkner
    Fachbereich Psychologie, WWU

    In recent years, Bayesian Statistics has gained a lot of popularity in the Natural and Social Sciences. Not only does it offer a consistent framework for probabilistic models, but it also provides researchers with an enormous modeling flexibility. It is this flexibility that allows creating statistical models tailored to the requirements of data, not the other way around. In this talk, I will present some general principles that guide the development, estimation, and interpretation of Bayesian statistical models. Among other things, I will talk about the choice of appropriate response distributions, the modeling of non-linear relationships as well as approaches to model comparisons in the Bayesian framework.

    Einladender:  O. Kamps


    Fluctuation response and large signal stability in power grids

    Dr. Frank Hellmann
    Potsdam Institute for Climate Impact Research, Potsdam

    In the complex network study of power grids, their dynamics are typically modeled as a Kuramoto model with inertia. In order to simplify the analysis it is often the case that losses are neglected, leading to symmetric effective network Laplacian in the stability and fluctuation analysis. I will present new results and analysis that show that losses, which appear as a phase shift parameters as in the Kuramoto-Sakaguchi type inertial oscillator models, dramatically alter the nonlinear and frequency characteristics of these systems.

    Einladender:  O. Kamps


    Solvent mediated interactions between nanostructures in suspension and the influence of solvophobicity

    Andy Archer

    When pairs of objects immersed in a liquid medium encounter one another there is, of course, a direct interaction between then, but there is also a solvent mediated interaction, which depends on the size difference between the objects and the solvent molecules and many other factors. This interaction can be attractive, repulsive or even both, at different separations and distances between the objects. If the forces between the objects are strongly attractive, then this leads to aggregation of the particles, so controlling carefully these forces are important in maintaining the particles in solution. When the surface of the objects are solvophobic, or have solvophobic patches, then this can lead to strong attractive forces. For example, hydrophobic forces such as these are important in determining protein folding in water and other molecular ordering in living systems. I will give an overview of some of the physics of solvent mediated interactions and show how such potentials can be calculated using classical density functional theory (DFT), focussing in particular on the influence of solvophobicity.

    Einladender: Prof. U. Thiele


    Machine Learning in Psychiatry - Promises, Problems, and Perspectives

    PD. Dr. Tim Hahn
    Klinik für Psychiatrie und Psychotherapie, WWU

    While our knowledge regarding the biological bases of psychiatric disorders has expanded massively in the last two decades, none of these findings have yet been translated into concrete clinical applications. One reason for this is that commonly conducted statistical analyses – while allowing for inference on the group level – do not enable predictions on the level of the individual. Another reason is the multifaceted nature of mental disorders which necessitates the integration of multiple sources of information when aiming to comprehensively represent relevant patient characteristics. Against this background, methods from the field of machine learning and multivariate pattern recognition have recently gained increasing attention, especially in the area of neuroimaging. The presentation gives an overview of the most recent developments in the emerging field of Predictive Analytics in Mental Health Research, providing concrete examples based on functional and structural Magnetic Resonance Imaging assessments, proteomic data as well as large-scale psychometric measures. Specifically, we will detail best practive project planning and common pitfalls when constructing models 1) predicting therapeutic response, 2) supporting differential diagnoses and 3) optimizing risk detection in the context of prevention and patient management. Finally, we will suggest specific guidelines for Deep Learning and Big Data approaches in multi-center research with the goal of providing the clinician with easy-to-use predictive algorithms and databases enabling early diagnosis and, ultimately, more efficient and successful treatment.

    Einladender: Dr. O. Kamps


    Formation and Spatial Localization of Phase Field Quasicrystals

    Priya Subramanian

    The dynamics of many physical systems often evolve to asymptotic states that exhibit spatial and temporal variations in their properties such as density, temperature, etc. Regular patterns such as graph paper and honeycombs look the same when moved by a basic unit and/or rotated by certain special angles. They possess both translational and rotational symmetries giving rise to discrete spatial Fourier transforms. In contrast, an aperiodic crystal displays long range order but no periodicity. Such quasicrystals lack the lattice symmetries of regular crystals, yet have discrete Fourier spectra. We look to understand the minimal mechanism which promotes the formation of such quasicrystalline structures arising in diverse soft matter systems such as dendritic-, star-, and block co-polymers using a phase field crystal model. Direct numerical simulations combined with weakly nonlinear analysis highlight the parameter values where the quasicrystals are the global minimum energy state and help determine the phase diagram. By locating parameter values where multiple patterned states possess the same free energy (Maxwell points), we obtain states where a patch of one type of pattern (for example, a quasicrystal) is present in the background of another (for example, the homogeneous liquid state) in the form of spatially localized dodecagonal (in 2D) and icosahedral (in 3D) quasicrystals. In two dimensions, we compute several families of spatially localized quasicrystals with dodecagonal structure and investigate their properties as a function of the system parameters. The presence of such metastable localized quasicrystals is significant as they affect the dynamics of the soft matter crystallization process.

    Einladender: Prof. U. Thiele


    Intermittent behavior across scales in biology

    Fernando Peruani
    Université de Nice Sophia Antipolis

    Intermittent behavior is observed in biological systems at all scales, from bacterial systems to sheep herds. First, I will discuss how Escherichia coli explores surfaces by alternating stop and moving phases. Specifically, I will show that a stochastic three behavioral state model is consistent with the empirical data. The model reveals that the stop frequency of bacteria is tuned at the optimal value that maximizes the diffusion coefficient. These results provide a new perspective on how evolution may have reshaped bacterial motility apparatus. Intermittent motion is also observed in sheep, where again a stochastic three behavioral state model provides a quantitative understanding of the empirical data. However, in sheep, individual transition rates depend on the behavioral state of other individuals and collective behaviors emerge. Specifically, I will show that small sheep herds display highly synchronized intermittent collective motion, with the herd behaving as a self-excitable system.

    Einladender: Prof. U. Thiele


    Computational Social Science

    Jun.-Prof. Dr. Annie Waldherr
    Institut für Kommunikationswissenschaft, WWU Münster

    Computational social science (CSS) is a growing interdisciplinary approach to study social processes using big data, advanced data analytics, and computer simulation. Besides just scaling traditional research approaches to massive data sets, some strands of CSS - such as network science and social simulation - also stress the interconnectedness and nonlinearity of social processes, embracing ideas from complexity science. After giving a broad overview on the main areas of current CSS, I will present an example of studying the public sphere as a complex system with agent-based modeling. In terms of complexity theory, the public sphere can be conceptualized as an arena where heterogeneous, autonomous and adaptive agents interact and self-organize. Multiple feedback mechanisms lead to the emergence of non-linear macro patterns such as news waves or opinion spirals. I show how the public sphere can be modeled as an agent-based system, and how the social simulation approach helps to develop theories of the public sphere, specifically by offering proof of concepts of existing theories, finding underlying mechanisms able to generate observed empirical patterns, and exploring possible futures.

    Einladender: Dr. O. Kamps


  • Colloquium Summer Term 2018


    A quantiative theory for transport equations

    Prof. Dr. Christian Seis
    Institut für Analysis und Numerik

    In this talk, I will review a new quantitative theory for transport equations with rough coefficients and discuss some of its main applications. More precisely, we consider examples in the context of fluid mixing and of numerical and stochastic approximations and investigate properties of solutions to the 2D Euler equations.



    A short course on Markov properties for physicists: Theory, numerical simulations and applications to biophysics

    Prof. Dr. Manuel Morillo Buzón
    Universidad de Sevilla

    The description of Markov Processes (MP) and their relevance to the physical
    sciences will be discussed. The random propagator and its density function will be
    the key ingredients to characterize several interesting types of MP:

    a) continuous MP and the Langevin description of Brownian motion
    b) jump continuous MP
    c) discrete jump MP and birth-death MP with applications to (bio)chemical kinetics.

    Issues like stability, stationary states and mean passage times between multistable
    systems will be addressed. The nonlinearity of realistic processes precludes their
    full analytical treatment. We will then discuss how the different types of problems
    can be numerically simulated.

    18.04.2018 (Mi.)    14 – 16 Uhr        SR 304, Wilhelm-Klemm-Straße 9
    20.04.2018 (Fr.)      14 – 16 Uhr       HS 404, Wilhelm-Klemm-Straße 9
    23.04.2018 (Mo.)   16 – 18 Uhr        HS 404, Wilhelm-Klemm-Straße 9
    24.04.2018 (Di.)     16:30 – 18 Uhr HS 404, Wilhelm-Klemm-Straße 9


    Predictive Modelling of Time Series in Neuroscience: The State Space Approach

    Dr. Andreas Galka
    Christian-Albrechts-University of Kiel, Institute for Medical Psychology and Medical Sociology

    Laith Hamid M.Sc.
    Christian-Albrechts-University of Kiel, Institute for Medical Psychology and Medical Sociology

    The last two decades have witnessed a major advancement of predictive modelling of time series data in Neuroscience. A key concept within this field is given by the methodology of Linear State Space (LSS) Modelling. In this talk we will review the basic concepts of LSS Modelling and discuss several practical applications, such as signal separation, noise reduction, artifact suppression, time series fusion and source analysis. While time series may have very different data dimensions and temporal sampling rates, LSS modelling provides a unifying framework for capturing the dynamics of the underlying neural processes. The parametric structure of LSS models enables the definition of "dynamical templates" which have been successfully applied to various filtering tasks. Additionally, in source analysis the generalization of the system model in LSS models enables region-specific modeling of dynamic processes in the brain. The generalization of the measurement equation allows for multimodal fusion. Compared to inverse solutions that ignore the temporal aspect in the data, dynamical inverse solutions based on LSS modelling may perform better in case of oscillatory activity, deep brain sources, presence of measurement noise and recordings with a small number of sensors.


    Joint work with Botond Cseke, Ramon Grima and Guido Sanguinetti.

    Using ideas form statistics for analysing (spatio-temporal) stochastic processes

    Dr. David Schnörr
    Theoretical Systems Biology group, Imperial College London

    Many systems in nature consist of stochastically interacting agents or particles. Stochastic processes have been widely used to model such systems, yet they are notoriously difficult to analyse. In this talk I will show how ideas from statistics can be used to tackle some challenging problems in the field of stochastic processes.
    In the first part, I will consider the problem of inference from experimental data for stochastic reaction-diffusion processes. I will show that multi-time distributions of such processes can be approximated by spatio-temporal Cox processes, a well-studied class of models from computational statistics. The resulting approximation allows us to naturally define an approximate likelihood, which can be efficiently optimised with respect to the kinetic parameters of the model.
    In the second part, we consider more general path properties of a certain class of stochastic processes. Specifically, we consider the problem of computing first-passage times for Markov jump processes, which are used to describe systems where the spatial locations of particles can be ignored. I will show that this important class of generally intractable problems can be exactly recast in terms of a Bayesian inference problem by introducing auxiliary observations. This leads us to derive an efficient approximation scheme to compute first-passage time distributions by solving a small, closed set of ordinary differential equations.


    Pattern selection through directional quenching

    Prof. Dr. Arnd Scheel
    University of Minnesota, School of Mathematics

    Interfaces or boundaries affect the formation of crystalline phases in sometimes quite dramatic ways. We are interested in examples where such interfaces arise through directional quenching and separate a striped phase from a uniform, non-crystalline state. Examples range from the alignment of convection roles in Benard convection to the robust patterning through presomites in limb formation, or the formation of helicoidal precipitates in recurrent precipitation. It turns out that growing the crystalline region leads to the selection of crystallographic parameters near the interface. We describe results for model problems such as Allen-Cahn, Cahn-Hilliard, and Swift-Hohenberg equations that quantify crystalline strain for small and large speeds and predict alignment parallel, perpendicular, or at oblique angles to the growth interface.


    Rethinking Nonlinear Dynamics - Self-organized Protein Pattern Formation

    Dr. Jacob Halatek
    Faculty of Physics, Statistical and Biological Physics, Ludwig-Maximilians-Universität Munich

    Protein pattern formation is essential for the spatial organization of many intracellular processes like cell division, flagellum positioning, and chemotaxis. More generally, these systems serve as model systems for self-organization, one of the core principles of life. We present a rigorous theoretical framework able to generalize and unify pattern formation for quantitative mass-conserving reaction-diffusion models in complex system geometries. Within this framework, separation of diffusive mass redistribution on the level of conserved species provides a general mathematical procedure to decompose complex reaction-diffusion systems into effectively distinct functional units. Our approach reveals the general underlying bifurcation scenarios and the influence of system geometry on pattern formation. We apply this general framework to a range of specific intracellular pattern forming protein networks, and show how it facilitates the identification of general self-organisation principles.


    Transport und Aufbewahrung bestrahlter Brennelemente in CASTOR®-Behältern

    Dr. Frank Jüttemann
    GNS Gesellschaft für Nuklear-Service mbH

    Die Präsentation vermittelt einen Überblick zur trockenen Zwischenlagerung bestrahlter Brennelemente in CASTOR® V Behältern. Für den Transport und die Zwischenlagerung solcher Brennelemente entwickelte GNS bereits vor mehr als 30 Jahren einen damals neuartigen Behältertyp, den CASTOR®. Die CASTOR®-Familie mit ihren unterschiedlichen, kontinuierlich weiterentwickelten Baureihen ist heute ein international bekanntes Markenzeichen und Synonym für nukleare Sicherheit, Zuverlässigkeit und Innovation. Die Präsentation gibt Einblicke zu den Aspekten bestrahlte Brennelemente, Regelwerksanforderungen, Behälterentwicklung, Behälterherstellung und zur Situation der trockenen Zwischenlagerung bestrahlter Brennelemente in Deutschland.


    Creation and collapse of an (almost) axisymmetric air cavity in water

    Prof. Dr. Devaraj van der Meer
    Physics of Fluids, University of Twente

    In this talk, we discuss an experiment in which we create (almost) axisymmetric air cavities by driving a metal disc through an initially quiescent water surface and observe their subsequent gravity-induced collapse. We will shed light onto the observed phenomena using an approach that combines (i) experiments, (ii) numerical simulations based on a boundary integral method, and (iii) a simplified model of the collapse of a slender cavity in terms of a modified Rayleigh equation. First, we find that the neck of the cavity closes in a singular manner and is described by a power-law with exponent 1/2 plus logarithmic corrections that depend on the aspect ratio of the cavity. Secondly, we observe that the liquid jet that is formed at the singularity is preceded by an air jet that becomes supersonic when the cavity neck diameter equals roughly a millimeter. And finally, we break axisymmetry using disks with a slight azimuthal harmonic disturbance of the rim and find that the cavity walls oscillate linearly during collapse, with nearly constant amplitude and increasing frequency. This behavior can be understood from a linear amplitude equation that fully explains even the most complex shape that was observed.


    Bubbles with Great Potential! Molecular Control of Foam Properties

    Prof. Dr. Björn Braunschweig
    Institute of Physical Chemistry, Westfälische Wilhelms-Universität Münster

    Foams are ubiquitous in our daily lives and in industrial applications. The vast number of possibilities to use foam in processes and products originate from their unique physico-chemical properties. Classically, surfactants, polyelectrolytes and proteins are used as molecular building blocks to stabilize aqueous foam. In order to formulate foam with specific properties, its structure must be controlled at the molecular level of the ubiquitous gas-fluid interfaces [1-3].
    Molecular-level characterizations of fluid interfaces are essential and help to reveal structure-property relations inside foam. Here, the molecular composition, molecular order and interactions such as electrostatics dominate, and thus must be addressed with molecular level probes that can provide access to both interfacial solvent and solute molecules. For that reason, we perform experiments on different hierarchical elements and at several length scales. The interfacial molecular structure and interactions are studied by using nonlinear optical spectroscopy, tensiometry and surface dilatational rheology, while measurements of disjoining pressure isotherms and foam analysis provide information on single foam films (air/liquid/air) and macroscopic foams (stability and structure). In this presentation, building blocks like macromolecules and surfactants are discussed and are shown to give rise to classical (consistent with DLVO theory) and non-classical foam stabilization mechanisms which are controlled by their self-organisation at the interface. The latter can be tailored by the choice of the electrolyte’s pH, ionic strength and composition.
    While the properties of most foams cannot be changed after their formation a new class of photo-switchable surfactants (arylazopyrazoles) and their 2D assemblies at the air-water interface offer the opportunity to render fluid interfaces as well as macroscopic foam responsive to light irradiation. This opens exciting new possibilities for foams such as self-healing capabilities or the possibility to adapt foam properties by photo-switching of the surfactants.

    Funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement No 638278) is gratefully acknowledged.

    [1] M. Schnurbus, L. Stricker, B.J. Ravoo and B. Braunschweig, Langmuir, 2018, under revision.
    [2] S. Streubel, F. Schulze-Zachau, E. Weißenborn and B. Braunschweig, J. Phys. Chem. C, 2017, 121, 27992
    [3] F. Schulze-Zachau and B. Braunschweig, Langmuir, 2017, 33, 3499

  • List of Previous Talks


    Winter Term 2017/2018



    Stability with respect to shocks

    Prof. Dr. Ulrike Feudel
    Universität Oldenburg
    Natural or technical systems possess often several possible stable states of operation. Linear stability theory is the appropriate tool to study the stability properties of such states with respect to small perturbations. However, in nature perturbations are not necessarily small but are finite in size. We discuss two different methods how to investigate the stability with respect to large perturbations such as single shocks. Both methods aim to determine the distance to the boundary of the basin of attraction or the edge of chaos, respectively. The first method determines the minimal destabilizing perturbation for large dynamical systems such as networks. Besides the size of this perturbations the method allows also to obtain the direction of this perturbation. We illustrate this method using pollinator networks in ecology and energy networks and identify relations between the topology of a network and its stability properties. The second method measures return times to a stable state at the edge of chaos. This is demonstrated for the transition from laminar to turbulent motion in a shear flow.


    Data Science als Berufsfeld für Physiker und Mathematiker

    Dr. Anton Daitche
    New Yorker
    In den letzten Jahren hat der Beruf Data Science enorm an Bedeutung gewonnen, mit einer stark steigenden Nachfrage. Aus meiner Sicht kann Data Science ein sehr interessantes Betätigungsfeld für Physiker und Mathematiker sein. Ich werde erläutern vorstellen, was Data Science und ins besondere Machine Learning ist, ein paar aktuelle Entwicklungen in dem Feld beleuchten und konkrete Anwendungsfälle vorstellen. Im Anschluss gebe ich ein paar Tips für Studenten und Doktoranden, die in dieses Feld einsteigen wollen.


    A Phase Field Model for Thin Elastic Structures with Topological Constraint

    Prof. Patrick Dondl
    Abteilung für Angewandte Mathematik, Albert-Ludwigs-Universität Freiburg

    With applications in the area of biological membranes in mind, we consider the problem of minimizing Willmore’s energy among the class of closed, connected surfaces with given surface area that are confined to a fixed container. Based on a phase field model for Willmore’s energy originally introduced by de Giorgi, we develop a technique to incorporate the connectedness constraint into a diffuse interface model of elastic membranes. Our approach uses a geodesic distance function associated to the phase field to discern different connected components of the support of the limiting mass measure. We obtain both a suitable compactness property for finite energy sequences as well as a Gamma-convergence result. Furthermore, we present computational evidence for the effectiveness of our technique. The main argument in our proof is based on a new, natural notion of convergence to describe phase fields in three dimensions.


    Nonlinear dynamics and time delays in engineering applications

    Dr. Andreas Otto
    Institut für Physik, TU Chemnitz
    Time delay effects appear in many dynamical systems. Often the delay effect is a result of a transport phenomena and feedback and the systems are nonlinear. In this talk we discuss general aspects of such systems, which can be often found in engineering. The relevant applications ranging from manufacturing processes, such as rolling and metal cutting over gasoline engines to traffic flow dynamics. After an introduction to the field of time delay systems, we will first focus on systems with state-dependent delays. We show that in many situations equivalent representations with constant delays exist, which are much easier to analyze. In a second part systems with multiple and distributed delays are studied and we discuss our recent results on systems with dissipative delays. Dissipative delays are a specific class of time-varying delays and may lead to a hitherto unknown type of chaotic behavior in nonlinear delay systems.


    Adaptive mesh-refinement for nonconforming finite element methods

    Prof. Dr. Mira Schedensack
    Angewandte Mathematik, WWU
    Non-conforming finite element methods lead to robust discretizations for almost  incompressible materials in solid mechanics or to pointwise divergence-free ansatz functions in fluid mechanics. If the exact solution is not smooth enough (e.g., if the underlying
    domain is not convex), finite element methods show suboptimal convergence rates. Adaptive mesh-refinement algorithms driven by error estimators automatically refine the mesh at the singularity. This talk introduces nonconforming finite element methods and adaptive mesh-refinement and shows optimal convergence rates of the algorithm for some problems.


    Thin film modelling of surfactant-driven biofilm spreading

    Sarah Trinschek (AG Thiele)

    The spreading of bacterial colonies at solid air interfaces hinges on physical processes connected to the properties of the involved interfaces. The production of surfactant molecules by the bacteria is one strategy that allows the bacterial colony to efficiently expand over a substrate. These surfactant molecules affect the surface tension which results in an increased wettability as well as in
    outward-pointing Marangoni fluxes that promote spreading. These fluxes may cause an instability of the circular colony shape and the subsequent formation of fingers. In this work, we study the front instability of bacterial colonies at solid-air interfaces induced by surfactant production in the framework of a passive hydrodynamic thin film model which is extended by bioactive terms. We show that the interplay between wettability and Marangoni fluxes determines the spreading dynamics and decides whether the colony can expand over the substrate. We observe four different types of spreading behaviour, namely, arrested and continuous spreading of circular colonies, slightly modulated front lines and the formation of pronounced fingers.

    Collective Cell Migration in Embryogenesis Follows the Laws of Wetting

    Bernhard Wallermeyer (AG Betz)

    Collective cell migration is a fundamental process during embryogenesis and its initial occurrence, called epiboly, is an excellent in vivo model to study the physical processes involved in collective cell movements that are key to understand organ formation, cancer invasion and wound healing. In zebrafish, epiboly starts with a cluster of cells at one pole of the spherical embryo. These cells are actively spreading in a continuous movement towards its other pole until they fully cover the yolk. Inspired by the physics of wetting we determine the contact angle between the cells and the yolk during epiboly. Similar to the case of a liquid drop on a surface one observes three interfaces that carry mechanical tension. Assuming that interfacial force balance holds during the quasi-static spreading process, we employ the physics of wetting to predict the temporal change of the contact angle. While the experimental
    values vary dramatically, the model allows us to rescale all measured contact angle dynamics onto a single master curve explaining the collective cell movement. Thus, we describe the fundamental and complex developmental mechanism at the onset of embryogenesis by only three main parameters: the offset tension strength 𝛼, the tension ratio 𝛿 and the rate of tension variation 𝜆.


    Branched Covering Surfaces

    Prof. Dr. Konrad Polthier
    Freie Universität Berlin, AG Mathematical Geometry Processing

    Multivalued functions and differential forms naturally lead to the concept of branched covering surfaces and more generally of branched covering manifolds in the spirit of Hermann Weyl's book "Die Idee der Riemannschen Fläche" from 1913. This talk will illustrate and discretize basic concepts of branched (simplicial) covering surfaces starting from complex analysis and surface theory up to their recent appearance in geometry processing algorithms and artistic mathematical designs.
    Applications will touch differential based surface modeling, image and geometry retargeting, global surface and volume remeshing, and novel weaved geometry representations with recent industrial applications.


    Interfacial turbulence and regularization in falling films

    Dmitri Tseluiko
    School of Mathematics, Loughborough University, UK
    We consider a liquid film flowing down an inclined wall that may be subjected to an additional external effects, such as an electric field. We analyse the Stokes-flow regime, using both a non-local long-wave model and the full system of governing equations. For an obtuse inclination angle and strong surface tension, the evolution of the interface is chaotic in space and time. However, a sufficiently strong electric field has a regularising effect, and the time-dependent solution evolves into an array of continuously interacting pulses, each of which resembles a single-hump solitary pulse. For an acute inclination angle and a sufficiently small supercritical value of the electric field, solitary-pulse solutions do not exist, and the time-dependent solution is instead a modulated array of short-wavelength waves. When the electric field is increased, the evolution of the interface first becomes chaotic, but then is regularised so that an array of pulses is generated. A coherent-structure theory for such pulses is developed and corroborated by numerical simulations.


    Turbulence and pattern formation in a model for active fluids

    Dr. Michael Wilczek
    MPISD Göttingen

    We consider a liquid film flowing down an inclined wall that may be subjected to an additional external effects, such as an electric field. We analyse the Stokes-flow regime, using both a non-local long-wave model and the full system of governing equations. For an obtuse inclination angle and strong surface tension, the evolution of the interface is chaotic in space and time. However, a sufficiently strong electric field has a regularising effect, and the time-dependent solution evolves into an array of continuously interacting pulses, each of which resembles a single-hump solitary pulse. For an acute inclination angle and a sufficiently small supercritical value of the electric field, solitary-pulse solutions do not exist, and the time-dependent solution is instead a modulated array of short-wavelength waves. When the electric field is increased, the evolution of the interface first becomes chaotic, but then is regularised so that an array of pulses is generated. A coherent-structure theory for such pulses is developed and corroborated by numerical simulations.


    Curve fitting on Riemannian manifolds

    Prof. Pierre-Antoine Absil
    Universitté Catholique de Louvain

    In this talk I will discuss curve fitting problems on manifolds. Manifolds of interest include the rotation group SO(3) (generation of rigid body motions from sample points), the set of 3x3 symmetric positive-definite matrices (interpolation of diffusion tensors) and the shape manifold (morphing). Ideally, we would like to find the curve that minimizes an energy function E defined as a weighted sum of (i) a sum-of-squares term penalizing the lack of fitting to the data points and (ii) a regularity term defined as the mean squared acceleration of the curve. The Euler-Lagrange necessary conditions for this problem are known to take the form of a fourth-order ordinary differential equation involving the curvature tensor of the manifold, which is in general hard to solve. Instead, we simplify the problem by restricting the set of admissible curves and by resorting to suboptimal Riemannian generalizations of Euclidean techniques.


    Branched flows in weakly scattering random media: from electronic transport to tsunami propagation

    Dr. Ragnar Fleischmann
    Max-Planck-Institut für Dynamik und Selbstorganisation, Göttingen

    Wave propagation in random media — this might sound abstract but is in fact very tangible and almost omnipresent in science and everyday life. Examples are wind driven ocean waves, but also light, sound, electrons, tsunamis and even earth quakes are waves that in a natural environment typically propagate through a complex medium. Due to its complexity, the medium is often best described as random with inherent correlations. Examples include the turbulent atmosphere, complex patterns of ocean currents or semiconductor crystals sprinkled with impurities. In recent years it has become clear that even very small fluctuations in a random medium, if they are correlated, lead to focussing of the waves in pronounced branch-like spatial structures and to heavy-tailed intensity distributions. These branches are closely connected with the occurrences of random caustics, i.e. singularities in the corresponding ray fields.
    I will give an overview over the phenomenon of branching and the statistical characteristics of branched flows, discussing examples from ballistic electron transport in semiconductors to the random focusing of tsunamis waves.


    Summer Term 2017

    How to tie a (linear optical) field into a knot

    Prof. Mark Dennis
    School of Physics, University of Bristol
    It is a challenging question to write down a function from real 3-dimensional space to the complex numbers such that the preimage if zero (say) is a given knot or link.  If, in addition, the function appears as a solution of some physically interesting partial differential equation, or minimizes some physically motivated functional, then the knotted field might be realisable in nature.  I will discuss our approach and (partial) solution this problem applied to such knotted fields in coherent optical fields (i.e. laser beams), but with applications to other systems such as knotted vorticity lines in fluids. If there is time, I will also describe how random fields (which model modes of chaotic wave systems) naturally contain a tangle of many knotted nodal lines.


    Nach dem Physikstudium in die KI-Industrie oder: Wie man die schlauen Maschinen endlich zum Arbeiten bringt

    Dr. Michael Köpf
    Cognotekt GmbH, Köln

    Die Ergebnisse jahrzehntelanger Grundlagenforschung im Bereich der Künstlichen Intelligenz schlagen sich aktuell in immer mehr Produkten und Anwendungen nieder. Mittlerweile ist KI in den Unternehmen angekommen und verändert durch die Automatisierung einfacher geistiger Tätigkeiten unsere Art zu arbeiten fundamental. Neben großen Playern wie Google, Amazon oder Baidu tragen auch mehr und mehr kleine bis mittelständische Firmen zu diesem Wandel bei. Physikerinnen und Physiker sind aufgrund der ihnen im Studium vermittelten Fähigkeiten besonders geeignet, diesen Umbruch mitzugestalten. Um ein Gefühl für dieses sehr moderne Berufsfeld zu vermitteln, werde ich anhand konkreter Beispiele aus der Versicherungsbranche die mit der Kommerzialisierung künstlicher Intelligenz verbundenen Herausforderungen vorstellen.


    Nonlinear dynamics of beating cilia and flagella: Swimming, steering, and synchronization

    Dr. Benjamin M. Friedrich
    TU Dresden, Center for Advancing Electronics Dresden (cfaed), Biological Algorithms Group

    Cilia and flagella represent a best-seller of nature: their regular bending waves propel cellular swimmers such as sperm cells and green alga in a liquid. Collections of these slender cell appendages can synchronize their beat to pump fluids inside human airways and brain ventricles effectively.
    In this talk, I will address the physics of flagellar swimming and how mechanical and chemical signals control these biological oscillators.
    In the first part, I will a present a theory of sperm chemotaxis, i.e. the directed navigation of flagellated sperm cells in response to signaling molecules released by the egg. We show how swimming along helical paths results in an effective navigation strategy that can cope with molecular shot noise of cellular concentration measurements. Thereby, spatial information about a concentration gradient becomes encoded in the phase of an oscillatory temporal signal perceived by the swimming cell along its helical path. This theory has recently been confirmed by experiments that track swimming sperm cells in three space dimension in artificial concentration fields of signaling molecules [1].
    In the second part, I will discuss flagellar synchronization as an emergent phenomenon in collections of several flagella, which arises from mutual a hydro-mechanical coupling. We present a theory of the beating flagellum as a noisy limit-cycle oscillator, which is fully calibrated by experimental data. In particular, we show using theory and experiment how external mechanical forces change speed and shape of the flagellar beat, or even stall the beat reversibly [2]. This flagellar load-response is key prerequisite for flagellar synchronization.
    We present a link between the efficiency of the flagellum to convert chemical energy into mechanical work and its ability to synchronize. Finally, we characterize the beating flagellum as a noisy oscillator, whose non-equilibrium fluctuations induce stochastic phase-slips in pairs of phase-locked flagella [3].

    [1] J.F. Jikeli et al.: Sperm navigation along helical paths in 3D chemoattractant landscapes, Nature Communications 6, 2015
    [2] G.S. Klindt, C. Ruloff, C. Wagner, B.M. Friedrich, Load-response of the flagellar beat, Phys. Rev. Lett. 117, 2016
    [3] R. Ma, G.S. Klindt, I.-H. Riedel-Kruse, F. Jülicher, B.M. Friedrich: Active phase and amplitude fluctuations of flagellar beating, Phys. Rev. Lett. 113, 2014


    Timing stability of quantum dot based semiconductor lasers

    Dr. Stefan Breuer
    Technische Universität Darmstadt, Angewandte Halbleiteroptik und Photonik

    Passively mode-locked semiconductor lasers based on nano-scale quantum dots offer access to sub-picosecond short optical pulses thanks to their broad spectral bandwidth and ultra-fast gain dynamics. Their small footprint, multi-Gigahertz repetition rates, direct modulation capability as well as their monolithic layout makes them promising candidates for optical clock distribution, high bit-rate optical time division multiplexing and compact microwave/millimeter-wave signal generation. Their intrinsic pulse train timing stability and fixed pulse repetition rate however can be limiting factors towards their widespread implementation into time-critical applications. Experimental concepts to improve the timing stability and to enable repetition rate agility are therefore demanded. This talk will start by reviewing timing stability fundamentals of semiconductor lasers. Then, a selection of experimental concepts to improve the timing stability will be described in detail. Next, by means of a reconfigurable laser layout, higher harmonics of the fundamental repetition rate can be generated leading to a substantially improved timing stability. Finally, experimental results will be discussed and explained in the framework of a time-domain description that is able to reproduce the timing stability improvement and repetition rate agility.


    On-chip generation of complex optical quantum states and their coherent control

    Dr. Michael Kues
    Institut national de la recherche scientifique, Varennes (Québec), Canada / Centre Energie Matériaux Télécommunications

    Entangled optical quantum states are essential towards solving questions in fundamental physics, and are at the heart of applications in quantum information science [1]. For advancing the research and development of quantum technologies, practical access to the generation and manipulation of complex photon states is required. Recently, integrated (on-chip) photonics has become a leading platform for the compact, cost-efficient, and stable generation and processing of optical quantum states [2]. However, on-chip sources are currently limited to basic two-dimensional (qubit) two-photon states.
    Within this presentation, I will show that integrated frequency combs (on-chip light sources with a broad spectrum of evenly-spaced frequency modes) based on high-Q nonlinear microring resonators can provide solutions towards scalable complex quantum state sources. Particularly, by using spontaneous four-wave mixing within the microring resonators, we demonstrate the generation of bi- and multi-photon entangled qubit states over a broad frequency comb spanning the telecommunications band, and control these states coherently to perform quantum interference measurements and tomographic reconstruction of their density matrix [3-5]. Moreover, we demonstrate the on-chip generation of entangled high-dimensional (quDit) states, where the photons are created in a coherent superposition of multiple pure frequency modes. In particular, we confirm the realization of a quantum system with at least one hundred dimensions. Furthermore, using off-the-shelf telecommunications components, we introduce a platform for the coherent manipulation and control of frequency-entangled quDit states.
    Our results suggest that microcavity-based entangled photon states and their coherent control using accessible telecommunications infrastructure can open up new venues for reaching the processing capabilities required for meaningful quantum information science.

    [1] Knill, E., Laflamme, R. & Milburn, G. J. “A scheme for efficient quantum computation with linear optics,” Nature 409, 46–52 (2001).
    [2] Tanzilli, S. et al. “On the genesis and evolution of integrated quantum optics,” Laser Photonics Rev. 6, 115–143 (2012).
    [3] C. Reimer, M. Kues, et al., “Integrated frequency comb source of heralded single photons,” Opt. Express. 22, 1023 (2014).
    [4] C. Reimer, M. Kues, et al., “Cross-polarized photon-pair generation and bi-chromatically pumped optical parametric oscillation on a chip,” Nat. Commun. 6, 8236 (2015).
    [5] C. Reimer, M. Kues, et al., “Generation of multiphoton entangled quantum states by means of integrated frequency combs,” Science 351, 1176 (2016).


    Structure formation in confinement: photonic balls and active granular rotors

    Prof. Dr. Michael Engel
    Friedrich-Alexander Universität Erlangen-Nürnberg, Institute for Multiscale Simulation

    Natural and biological systems achieve emergent behavior with elementary building blocks. Research in my group focuses on modeling structure formation processes in soft and hard condensed matter. In this talk, I discuss two joint computational-experimental works that have in common that they concern structure formation of particles in confinement. (1) A binary mixture of 3D-printed macroscopic rotors with opposite sense of rotation is excited on an electromagnetic shaker in circular confinement. Phase-separation reminiscent of spinodal decomposition is observed. Evolution of the domain size is compared to two-dimensional Langevin dynamics simulations. (2) Droplet-based microfluidics creates homogeneous emulsion droplets as sources for defined spherical confinement. We observe a discrete series of multiply twinned colloidal clusters with icosahedral symmetry. To understand and explain the formation of the clusters, we test a geometric model and extract extremal principles.


    Network robustness and the impact of transmission line failures

    Prof. Dr. Dirk Witthaut
    Forschungszentrum Jülich GmbH, Institut für Energie- und Klimaforschung, Systemforschung und Technologische Entwicklung (IEK-STE)

    The robust operation of physical distribution and supply networks is fundamental for economy, industry, and our daily life. For instance, a reliable supply of electric power fundamentally underlies the function of most of our technical infrastructure. In periods of high loads, the breakdown of a single element of the power grid can cause a global cascade of failures implying large-scale outages with potentially catastrophic consequences.
    In this talk I will review the theory of transmission line outages and the robustness of networks. Once a line in the network fails, the flow must be rerouted over alternative pathways. To assess the robustness of a network we must understand where flows are rerouted and quantify how much capacity is available for this task. I will discuss the mathematical formulation of the problem, present some recent results, and demonstrate some illuminating connections to other fields of theoretical physics and applied math.


    Spontaneous and Coherent Raman microscopy for biomedical applications

    Dr. Cees Otto
    University of Twente, Faculty of Science and Technology, Medical Cell BioPhysics (MCBP)

    Raman microscopy is well known as a powerful method for the study of molecular changes and processes. In biomedical applications the chemical information is usually extremely complex.
    The laser-based implementation of vibrational spectroscopy that underlies modern spontaneous and Coherent Raman microscopy however enables a broad range of applications also in the biomedical field. In this presentation an introduction to various forms of Raman microscopy will be presented with applications on cells and tissues.


    Theory, structure and experimental justification of the metal/electrolyte interface

    Dr. Manuel Landstorfer
    Weierstrass Institute for Applied Analysis and Stochastics (WIAS), Berlin

    Various types of modeling approaches are used to investigate the structure of the metal/electrolyte interface and its behaviour due to adsorption, intercalation and other surface effects [1]. While atomistic resolved models address reaction mechanisms, continuum models are the foundation for a comparison to voltage- or current-controlled experiments.
    In this talk I will provide insight to our model framework which explicitly accounts for solvation and adsorption effects. We will show that this model is the very basis for a qualitative and quantitative understanding of the capacitive behaviour of a variety of electrolytes. We will further show that our approach is also the very basis for a model based understanding of cyclic voltammetry and thus a key tool for analytical electrochemistry.
    Most common continuum models essentially relay on simplified Poisson–Boltzmann equations to model the electrochemical double layer. However, it is long known [2] that this equation is not able to predict the capacitive behaviour of electrochemical interfaces. We showed that this is due to the negligence of the incompressibility [3] and solvation effects [4] of electrolytic solutions. Our new model essentially leads to a coupled Poisson–momentum equation system and is able to predict the non-linear behaviour of the double layer capacity over the whole potential range. Additionally, our surface mixture theory [5] accounts for adsorption on the metal surface and for partial solvation, thus leading to an overall precise model which is able to predict the unsymmetric capacity curve of strongly adsorbing ions (i.e. NaF or NaClO4[5]). Based on this, we are able to predict the thermodynamic structure of the space charge layer at the metal/electrolyte interface, which is in full agreement to experimental data.


    [1] M. Landstorfer and T. Jacob, Chem. Soc. Rev., 2013, 42, pp 3234–3252.
    [2] J. Bockris, A. Reddy and M. Gamboa-Aldeco, Modern Electrochemistry 2A: Fundamentals of Electrodics, Springer, 2001, vol. 2.
    [3] W. Dreyer, C. Guhlke and R. Müller, Phys. Chem. Chem. Phys., 2013, 15 , pp 7075–7086.
    [4] W. Dreyer, C. Guhlke and M. Landstorfer, Electrochemistry Communications, 2014, 43 , pp 75 – 78.
    [5] W. Dreyer, C. Guhlke and M. Landstorfer, Electrochimica Acta, 2016, 201, 187 – 219.
    [6] G. Valette, J. of Electroanal. Chemistry and Interfacial Electrochemistry, 1981, 122 , pp 285 – 297.
    [7] G. Valette, Journal of Electroanalytical Chemistry and Interfacial Electrochemistry, 1982, 138, pp 37 – 54.


    Static and Dynamic Functional Brain Connectivity at Sensor and Source level: Evidences from EEG-MEG Group Analysis

    Dr. Stavros Dimitriadis
    Aristotle University of Thessaloniki

    The human brain can be modelled as a complex networked structure with brain regions as individual nodes and their anatomical/functional links as edges. Functional brain networks are constructed by first extracting weighted connectivity matrices, and then thresholding them to minimize the noise level. Different methods have been used to estimate the dependency values between the nodes The adaptation of both bivariate (mutual information) and multivariate (Granger causality) connectivity estimators to quantify the synchronization between multichannel recordings yields a fully connected, weighted, (a)symmetric functional connectivity graph (FCG), representing the associations among all brain areas. The aforementioned procedure leads to an extremely dense network of tens up to a few hundreds of weights. Therefore, this FCG must be filtered out so that the “true” connectivity pattern can emerge.  For that reason, statistical filtering based on surrogates analysis and also topological filtering based on the maximization of information flow in the network under the constraint of the wiring cost (Dimitriadis et al., 2017) should be adopted to get a subject and condition specific functional connectivity pattern.
    The whole methodology relies on data-driven techniques without using a priori information for the subjects e.g. labels. Subject-specific approaches increase the reproducibility of the dataset avoiding optimizing it independently for each study. Complementary, one can add more subjects to the original cohort without re-optimizing the parameters, an approach that can lead to controversy findings to the original study.
    I will demonstrate how different connectivity estimators in both static and dynamic functional connectivity with the incorporation of arbitrary statistical and topological filtering schemes can give contradictory results. The selection of appropriate surrogates and data-driven topological filtering scheme will give more reproducible and stable results. The whole analysis will focus on electro/magneto-encephalography (EEG/MEG) at resting-state in normal populations and in mild cognitive impairments (MCI) subjects. First evidences of similarities of connectivity patterns between sensor and source-level will be also demonstrated.


    [1] Dimitriadis SI et al., . Topological Filtering of Dynamic Functional Brain Networks Unfolds Informative Chronnectomics: A Novel Data-Driven Thresholding Scheme Based on Orthogonal Minimal Spanning Trees (OMSTs). Front. Neuroinform., 26 April 2017 | https://doi.org/10.3389/fninf.2017.00028


    Winter Term 2016/2017

    Variational Mean Field Games

    Prof. Dr. Filippo Santambrogio, Laboratoire de Mathématiques d'Orsay, Université Paris-Sud

    I will give a brief introduction to the the emerging topic of Mean Field Games, introduced by J-M Lasry and P-L Lions some years ago as a model for the Nash equilibrium of a population of agents each selecting his own evolution, taking into account the density of the other agents that he meets, in the form of a congestion charge. This gives rise to a coupled system of PDEs, a continuity equation where the density moves according to the gradient of a value function, and a Hamilton-Jacobi equation solved by the value function, where the density also appears.
    I will mainly deal with the case where this equilibrium problem may be seen as optimality conditions of a convex variational problem, and give the main results in this framework. I will also present some recent regularity results which allow to give a precise mathematical meaning to the equilibrium condition. At the end of the talk I will present an interesting variant, where the congestion cost is replaced by a capacity constraint. In this case an extra unknown appears, which plays both the role of a pressure in the fluid mechanics point of view, and of a price in the economical interpretation.


    Nonlinear dynamics of bacterial and active colloidal suspensions

    Prof. Dr. Hartmut Löwen, Institut für Theoretische Physik II: Weiche Materie, Heinrich-Heine-Universität Düsseldorf

    This talk addresses the nonlinear dynamical aspects of active matter for both biological and
    artificial colloidal microswimmers. 
    [1]. A number of collective phenomena in active matter will be proposed. In detail, we shall discuss active turbulence in bacterial systems [2] and its ability to steer the transport of shuttles [3]. Then we turn to kinetic phase separation in artificial swimmers [4]. The late time kinetics of phase separation can be mapped on the Cahn-Hilliard model [5]. Moreover we put forward the concept of negative interfacial tension [6] in nonequilibrium phase-separated swimmer systems and that of the swim pressure at system boundaries [7].


    [1] For a recent review, see: C. Bechinger, R. di Leonardo, H. Lowen, C. Reichhardt, G. Volpe, G. Volpe,
      Active Brownian motion in complex and crowded environments, arXiv: 1602.00081

    [2] H. H. Wensink, J. Dunkel, S. Heidenreich, K. Drescher,  R. E. Goldstein,  H. Lowen, J. M. Yeomans, PNAS  109, 14308  (2012).

    [3] A. Kaiser, A. Peshkow, A. Sokolov, B. ten Hagen, H. Lowen, I. S. Aranson, Physical Review Letters 112, 158101  (2014).

    [4] I. Buttinoni, J. Bialke, F. Kummel, H. Lowen, C. Bechinger, T. Speck, Physical Review Letters 110, 238301  (2013).

    [5] T. Speck, J. Bialke, A. M. Menzel, H. Lowen, Physical Review Letters 112, 218304 (2014).

    [6] J. Bialk'e, J. T. Siebert, H. Lowen, T. Speck,Physical Review Letters  115, 098301  (2015).

    [7] F. Smallenburg, H. Lowen, Physical Review E 92, 032304 (2015).


    Physical modeling of cellular motility

    PD Dr. Falko Ziebert, Physikalisches Institut Fakultät für Mathematik und Physik, Albert-Ludwigs-Universität Freiburg

    Substrate-based crawling motility of eukaryotic cells is essential for many biological functions, both in developing and mature organisms. Motility dysfunctions are involved in several life-threatening pathologies such as cancer and metastasis. Motile cells are also a natural realization of active, self-propelled ‘particles’, a popular research topic in nonequilibrium physics. Finally, from the materials perspective, assemblies of motile cells and evolving tissues constitute a class of adaptive self-healing materials that respond to the topography, elasticity, and surface chemistry of the environment and react to external stimuli.
    Although a comprehensive understanding of substrate-based cell motility remains elusive, progress has been achieved recently in its modeling on the whole cell level. I will survey our recent advances in computational approaches to cell movement on structured substrates (with modulated adhesion or stiffness), as well as on collective cell migration.

    Ziebert, F. & Aranson, I. S. Effects of adhesion dynamics and substrate compliance on the shape and motility of crawling cells. PLoS ONE 8, e64511 (2013).

    Löber, J., Ziebert, F. & Aranson, I. S. Modeling crawling cell movement on soft engineered substrates. Soft Matt. 10, 1365–1373 (2014).

    Löber, J., Ziebert, F. & Aranson, I. S. Collisions of deformable cells lead to collective migration. Sci. Rep. 5, 9172 (2015).

    Ziebert, F., Löber, J. & Aranson, I. S. Macroscopic model of substrate-based cell motility. in Physical Models of Cell motility, Ed. I. S. Aranson, Springer (Switzerland) 1–67 (2016).


    The Fibonacci family of dynamical universality classes

    Prof. Dr. Gunter M. Schütz, Theoretical Soft Matter and Biophysics, Forschungszentrum Jülich GmbH

    We use the theory of nonlinear fluctuating hydrodynamics to study stochastic transport far from thermal equilibrium in terms of the dynamical structure function which is universal at low frequencies and for large times and which encodes whether transport is diffusive or anomalous. For generic quasi one-dimensional systems we predict [1] that transport of mass, energy and other locally conserved quantities is governed by mode-dependent dynamical universality classes with dynamical exponents z which are Kepler ratios of neighboring Fibonacci numbers, starting with z = 2 (corresponding to a diffusive mode) or z = 3/2 (Kardar-Parisi-Zhang (KPZ) mode). If neither a diffusive nor a KPZ mode are present, all modes have as dynamical exponent the golden mean z=(1+\sqrt 5)/2. The universal scaling functions of the higher Fibonacci modes are Lévy distributions. The theoretical predictions are confirmed by Monte-Carlo simulations of a three-lane asymmetric simple exclusion process.

    [1] V. Popkov, A. Schadschneider, J. Schmidt, and G. M. Schütz, PNAS Early Edition (2015), www.pnas.org/cgi/doi/10.1073/pnas.1512261112


    Asymptotic homogenization for fluid and drug transport in vascularized tumors: theory and applications

    Dr. Raimondo Penta, Departamento de Mecanica de los Medios Continuos y T. Estructuras,
    E.T.S. de caminos, canales y puertos, Universidad Politecnica de Madrid

    A vascularized tumor is characterized by a sharp length scale separation between the intercapillary distance (microscale) and the average tumor size (macroscale). Application of the asymptotic homogenization technique leads to the macroscale double Darcy, double advection-diffusion-reaction type model for fluid and drug transport in the tumor, vessels, and across the capillaries’ walls. We recover microscopic information solving cell problems representative of the microvessels’ geometries. We present the theory [1], and applications concerning the role of the vessels’ geometry on blood convection [2] and diffusion/consumption of anticancer drugs [3].

    [1] R. Penta, D. Ambrosi, and A. Quarteroni. Multiscale homogenization for fluid and drug transport in vascularized malignant tissues. Mathematical Models and Methods in Applied Sciences, 25(1):79–108, 2015.

    [2] R. Penta and D. Ambrosi. The role of the microvascular tortuosity in tumor transport phenomena. Journal of theoretical biology, 364:80–97, 2015.

    [3] P.Mascheroni and R.Penta. The role of the angiogenic network structure on diffusion and consumption of anti-cancer drugs, Int. J. Numer. Meth. Biomed. Engng, submitted (2016).


    Bending, Buckling and Bifurcations

    Prof. Dr. Andrew Hazel, School of Mathematics,The University of Manchester

    Quantifying and controlling the buckling and bending of solid materials has applications from molecular scales (DNA) to large-scale natural and artificial structures. In general, buckling can be interpreted as a consequence of bifurcation phenomena in nonlinear systems. In this talk, I shall present results from our recent investigation of a new buckling mode that can arise after introducing internal structure, a line of holes, into a simple column [Soft Matter (2016), 12, pp 7112--7118]. This simple geometric modification introduces a smaller lengthscale into the system and has a profound effect on the solution structure, which becomes increasing complex as the number of holes increases.



    Controlling caustics: Weaving catastrophic structures in Airy-, Pearcey-, and swallowtail beams

    Alessandro Zannotti, AG Denz, Institut für Angewandte Physik

    Catastrophe science is a branch of bifurcation theory in the study of dynamical systems; it is also a special case of more general singularity physics. Bifurcation theory studies and classifies phenomena characterized by sudden shifts in behaviour arising from small changes in circumstances, analysing how the qualitative nature of solutions depends on control parameters. In optics, these dramatic changes manifest as geometrically stable caustics, which, as natural phenomena, are associated with the arcs close to rainbows, or may occur as high-intensity networks on the floor of shallow waters. Similar to their formation behind refractive index lenses with imperfections, the formation of corresponding structures has been observed for numerous kinds of lenses with importance in optics, astrophysics and surface analytics.
    The most prominent representative of catastrophe that manifests as wave package is the fold catastrophe that has been realized as paraxial Airy beam, and shows a form-invariant and accelerated propagation. Here, we present higher-order catastrophes as paraxial light, discuss the mapping of corresponding cusp and swallowtail catastrophes to these beams and demonstrate their unique propagation effects. Their often auto-focusing propagation of curved trajectories is well suited to optically induce photonic structures, waveguides to nonlinear photosensitive materials.

    Delay-induced Dynamics in a Swift-Hohenberg Model with Spatial Inhomogeneities

    Felix Tabbert, AG Thiele, Institut für Theoretische Physik

    We study pattern formation in a Swift-Hohenberg model that, aside from many other applications, describes the evolution of the electric field envelope in the transverse plane of a bistable optical cavity pumped by an external injection beam.
    Typical solutions of the Swift-Hohenberg equation are homogeneous, periodic and localized states. Here, we focus on the destabilization of stable localized solutions by time-delayed feedback, which can be implemented by using an external cavity in optical applications. We show that by varying the delay time and the delay strength, one can induce a variety of different dynamics including the drift or the annihilation of the localized solutions as well as travelling wave solutions.
    A special focus lies on the introduction of spatial inhomogeneities into the system, e.g., by introducing a spatially inhomogeneous injection beam, which changes the delay-induced dynamics drastically. We report on pinned localized solutions, oscillating solutions and depinning due to timedelayed feedback. The competition of a pinning inhomogeneity and a destabilizing time-delayed feedback is studied both analytically and numerically.


    Rational design of stimuli-responsive nanoreactors

    Prof. Dr. Joe Dzubiella, HU Berlin und Helmholtz Zentrum Berlin

    The catalysis by metal nanoparticles is one of the fastest growing fields in nanoscience. However, the optimal control of catalytic activity and selectivity in nanoparticle catalysis remains a grand scientific challenge. Here, we describe our ongoing efforts how to theoretically derive design rules for the optimization of nanoparticle catalysis (in the fluid phase) by means of thermosensitive yolk-shell and core-shell carrier systems [1-3]. In the latter, nanoparticles are stabilized in solution by an encapsulating, thermosensitive hydrogel shell. The latter contains and shelters the reaction. The physicochemical properties, in particular the permeability, of this polymeric 'nanogate' react to stimuli in the environment and thus permit the reactant transport and with that the catalytic reaction to be switched and tuned, e.g., by the temperature [1-3], salt concentration, or solvent composition. Hence, the novel hybrid character of these emerging 'nanoreactors' opens up unprecedented ways for the control of nanocatalysis due to new designable degrees of freedom, if theoretical understanding and rational design principles are available.


    [1] S. Wu, J. Dzubiella, J. Kaiser, M. Drechsler, X. Guo, M. Ballauff, Y. Lu, Angewandte Chemie 51, 2229 (2012).

    [2] P. Herves, M. Perez-Lorenzo, L. M. Liz-Marzan, J. Dzubiella. Y. Lu, and M. Ballauff, Chem. Soc. Rev. 41, 5577 (2012).

    [3] S. Angioletti-Uberti, Y. Lu, M. Ballauff, and J. Dzubiella, J. Phys. Chem. C 119, 15723 (2015).


    Temporal Localized Structures and Light Bullets in Passively mode-locked Lasers

    Dr. Julien Javaloyes, Departament de Fisica, Edifici Mateu Orfila, Universitat de les Illes Baleares

    Localized structures (LS) are nonlinear states of dissipative extended systems characterized by a correlation range much shorter than the size of the system, thus allowing for individual addressing. They appear ubiquitously in nature and they are very appealing in optical systems for applications to information processing, especially in semiconductor lasers which are fast, scalable and cheap devices. We investigate the relationship between passive mode-locking and the formation of temporal localized structures in the output intensity of a laser coupled to a saturable absorber, in the framework of time delayed dynamical systems. We present experimental and theoretical evidences [1] regarding how the mode-locked pulses transform into lasing localized structures, allowing for individual addressing and arbitrary low repetition rates. Our analysis reveals that this occurs when i) the cavity round-trip is much larger than the slowest medium timescale, namely the gain recovery time and ii) the mode-locked solution coexists with the stable off solution. These conditions enable the coexistence of a large quantity of stable solutions, each of them being characterized by a different number of pulses per round-trip and with different arrangements. A modulation of the bias current allows controlling the number and the location of the pulses traveling within the cavity [2, 3]. These results formed the basis for a very recent prediction [4] as we theoretically demonstrated the existence of three dimensional dissipative localized structures in the output of a broad area passively mode-locked laser coupled to a saturable absorber in self-imaging conditions. These phase invariant light bullets are individually addressable and can be envisioned for three dimensional optical information storage. An effective theory provides for an intuitive picture and allows to relate their formation to static auto-solitons.

    [1] M. Marconi, J. Javaloyes, S. Balle, and M. Giudici. How lasing localized structures evolve out of passive mode locking. Phys. Rev. Lett., 112:223901, Jun 2014.

    [2] M. Marconi, J. Javaloyes, P. Camelin, D.C. Gonzalez, S. Balle, and M. Giudici. Control and generation of localized pulses in passively mode-locked semiconductor lasers. Selected Topics in Quantum Electronics, IEEE Journal of, 21(6):1-10, Nov 2015.

    [3] J. Javaloyes, P. Camelin, M. Marconi, and M. Giudici. Dynamics of localized structures in systems with broken parity symmetry. Phys. Rev. Lett., 116:133901, Mar 2016.

    [4] J. Javaloyes. Cavity light bullets in passively mode-locked semiconductor lasers. Phys. Rev. Lett., 116:043901, Jan 2016.


    Inhomogeneous Boltzmann-type equations modelling opinion leadership and political segregation

    Dr. Bertram Düring, Department of Mathematics, University of Sussex

    In recent years different kinetic models to describe opinion formation have been proposed. Such models successfully use mathematical tools from statistical mechanics to describe the behaviour of a large number of interacting individuals in a society. This leads to generalizations of the classical Boltzmann equation for gas dynamics. Most approaches in the literature assume a homogeneous society. To model additional sociologic effects in real societies, e.g. the influence opinion leaders, one needs to consider inhomogeneous models.
    In this talk we discuss such kinetic models for opinion formation, where the opinion formation process depends on an additional independent variable, e.g. a leadership or a spatial variable. More specifically, we consider:
    (i) opinion dynamics under the effect of opinion leadership, where each individual is characterised not only by its opinion, but also by another independent variable which quantifies leadership qualities;
    (ii) opinion dynamics modelling political segregation in `The Big Sort', a phenomenon that US citizens increasingly prefer to live in neighbourhoods with politically like-minded individuals. Based on microscopic opinion consensus dynamics such models lead to inhomogeneous Boltzmann-type equations for the opinion distribution. We derive macroscopic Fokker-Planck-type equations in a quasi-invariant opinion limit and present results of numerical experiments.