>> DIEP seminars and talks - archives 2021
Below you can find more information on the DIEP seminars of 2021.
>> DIEP Seminar: Bjarke F. Nielsen (U. Roskilde)
Heterogeneous disease transmission - superspreading and the case for statistical physics | 9th of December 2021
​
The transmission pattern of SARS-CoV-2 has proven to be very heterogeneous, with a tendency towards superspreading. So much so that it has been estimated that just 10% of infected individuals give rise to 80% of new cases. This finding is surprisingly robust and has been reproduced by several methods, including studies based on contact tracing, aggregated incidence data and phylodynamics. Recent mathematical models have shown that this feature has profound - and sometimes positive - implications for the effectiveness of some non pharmaceutical interventions. Capturing the phenomenon requires concepts and methods from statistical physics, such as agent-based modelling and the distinction between quenched and annealed noise, and would not be possible in traditional, compartmental homogeneous-mixing models of disease spread. In a broad sense, these results highlight the importance of taking heterogeneity into account and provide a strong argument for the introduction of novel, statistical physics-inspired models into infectious disease modelling. Lastly, I will comment on how transmission heterogeneity and interventions can interact with pathogen evolution and the emergence of new variants.
>> DIEP Seminar: Tomas Veloz (Vrije Universiteit Brussel)
Reaction Networks and Evolutionary Game Theory | 2nd of December 2021
​
Mathematical approaches in systems biology are increasingly applied beyond the scope of biology. Particularly, reaction networks have been suggested as an alternative way to model systems of a general kind, and particularly social phenomena. In this latter “socio-chemical metaphor” molecular species play the role of agents’ decisions and their outcomes, and chemical reactions play the role of interactions among these decisions. From here, it is possible to study the dynamical properties of social systems using standard tools of biochemical modelling. In this talk we show how reaction networks can model systems that are usually studied via evolutionary game theory.
We illustrate our framework by modeling the repeated prisoners’ dilemma. We further develop a model considering the interaction among Tit for Tat and Defector agents.
​
We will discuss the strengths and weaknesses of the approach as well as its potential to produce new insights in classical problems such as the emergence of goal-directedness and the evolution of cooperation.
>> DIEP Seminar: Physics Nobel Prize 2021
An invitation to the work of Hasselmann, Manabe and Parisi | 25th of November 2021
​
The Nobel Prize in Physics 2021 was awarded "for groundbreaking contributions to our understanding of complex systems" with one half jointly to Syukuro Manabe and Klaus Hasselmann "for the physical modelling of Earth's climate, quantifying variability and reliably predicting global warming" and the other half to Giorgio Parisi "for the discovery of the interplay of disorder and fluctuations in physical systems from atomic to planetary scales."
The Dutch Institute for Emergent Phenomena invites you to attend a series of short lectures by John Mydosh, Edan Lerner, Elisabetta Pallante, Luca Giomi and Daan Crommelin on aspects of the Nobel Prize 2021 that cover many of the contributions that Hasselmann, Manabe and Parisi made to spin glasses, particle physics, flocking and climate models. The event takes place online. To receive the Zoom link before the event register below.
​
>> DIEP Seminar: Han van der Maas (U. Amsterdam)
Cascading transition in psychology | 18th of November 2021
​
Tipping points or phase transitions separate stable states in psycho-social systems. Examples are quitting smoking, radicalization, and dropping-out of school. Two knowledge gaps prevent our ability to predict and control these tipping points. First, we miss explanatory mathematical models of such non-linear processes. Second, we ignore the multilevel character of psycho-social transitions. I contend that important changes in many psycho-social systems are cascading transitions, where individual transitions trigger or are triggered by social transitions. The cascade of radicalization of individuals in the context of political polarization in societies is an example of such a multilevel process. Being able to predict and control cascading transitions in psycho-social systems would be a major scientific breakthrough.
As an expert of complex systems research in the behavioural and social sciences, with an extensive track record in studying single level psychological transition processes (e.g., in perception, sleep, disorders, cognition, and attitudes), my key objective is to develop a novel and broadly applicable methodology to study multilevel cascading transitions in psycho-social systems. This methodology comprises theory construction in the form of mathematical modelling and innovative empirical analyses. I will do so by studying three important examples: a) opinion change from individuals to populations and back, b) learning, where progression and drop-out are embedded in collective processes, c) addiction, where transitions to addiction or abstinence within individuals are part of cascading epidemiological changes of substance use in populations.
>> DIEP Seminar: Mari Carmen Banuls (Max Planck Institute of Quantum Optics)
Tensor networks for classical systems | 11th of November 2021
​
Tensor network states (TNS) are a very useful tool for the study of correlated many-body systems. In the context of quantum many-body problems, they provide precise and efficient methods to explore thermal equilibrium and ground states. And it is also possible to use them, to some extent, for the study of out of equilibrium dynamics.
TNS can also be applied to classical problems. One the one hand, it is possible to compute partition functions and observables in equilibrium. Furthermore, they can be combined with Montecarlo techniques to sample from a Boltzmann distribution for a wide variety of models. On the other hand, also dynamical scenarios can be studied with these techniques. In particular, we have recently shown how they can be used for studying large deviation functions and sampling rare trajectories in constrained stochastic models.
>> DIEP Seminar: Hugo Touchette (Stellenbosch University)
The large deviation approach to statistical physics | 4th of November 2021
​
I will give in this talk a basic overview of the theory of large deviations and of its applications in statistical physics. In the first part, I will discuss the basics of this theory and its historical sources, which can be traced back in mathematics to Cramer (1938), Sanov (1960) and Varadhan (1970s) and, on the physics side, to Einstein (1910) and Boltzmann (1877). In the second part, I will discuss how the theory has been applied in recent years to study various equilibrium, nonequilibrium, and complex systems, such as interacting particle systems, turbulent flows, and random graphs, among many examples.
_edited.jpg)
>> DIEP Seminar: Wolfram Barfuss (U. Tübingen)
Collective Learning Dynamics | 28th of October 2021
​
Collective learning will become of vital importance not only in genuine multi-agent systems, such as mobility, swarm robotics, or public infrastructure, it is also crucial to consider when applied machine learning systems are changing the very same data they have been trained upon. The question of an adequate theoretical foundation for multi-agent learning, however, remains unanswered. I will show how techniques from evolutionary game theory and statistical mechanics can provide an improved understanding of the emerging collective learning dynamics in changing environments.
>> DIEP Seminar: Matthijs van Veelen (U. Amsterdam)
The evolution of morality and the role of commitment | 21st of October 2021
​
A considerable share of the literature on the evolution of human cooperation considers the question why we have not evolved to play the Nash equilibrium in prisoners’ dilemmas or public goods games. In order to understand human morality and pro-social behaviour, we suggest that it would actually be more informative to investigate why we have not evolved to play the subgame perfect Nash equilibrium in sequential games, such as the ultimatum game and the trust game. The ‘rationally irrational’ behaviour that can evolve in such games gives a much better match with actual human behaviour, including elements of morality such as honesty, responsibility and sincerity, as well as the more hostile aspects of human nature, such as anger and vengefulness. The mechanism at work here is commitment, which does not need population structure, nor does it need interactions to be repeated. We argue that this shift in focus can not only help explain why humans have evolved to know wrong from right, but also why other animals, with similar population structures and similar rates of repetition, have not evolved similar moral sentiments. The suggestion that the evolutionary function of morality is to help us commit to otherwise irrational behaviour stems from the work of Robert Frank (Passions within reason: The strategic role of the emotions, 1988), which has played a surprisingly modest role in the scientific debate to date.
_edited.jpg)
>> DIEP Seminar: Justus Uitermark (U. Amsterdam)
Beyond complexity | 14th of October 2021
​
Recent years have seen the emergence of a new interdisciplinary field for the study of social life: computational social science. Computational social scientists often draw upon methods and concepts developed within the natural sciences to examine patterns in social processes and identify underlying mechanisms. In this presentation, I document my engagement with computational social science, asking the overarching question how social life is different from the complex systems commonly studied in the natural sciences and what implications this should have for ontology and epistemology. I suggest that the complexity perspective may be useful but has difficulty incorporating quintessentially social phenomenon like meaning and power. By way of illustration, I discuss research I have done with computational scholars on social movements, public debates, cities, and the diffusion of scientific ideas.
Justus Uitermark is a sociologist and geographer who studies how different kinds of – online and offline – environments shape political and cultural conflict. He currently works as a professor of urban geography at the University of Amsterdam and is affiliated with the UvA’s Center for Urban Studies as well as the Amsterdam Institute for Social Science Research.
>> DIEP Seminar: Alexandru Baltag (U. Amsterdam)
Group (Ir)Rationality: can logic help? | 7th of October 2021
​
I present some applications of logical methods to the study of emergent phenomena in groups of `agents', capable of reflection, communication, reasoning, argumentation etc.The main focus is on the understanding of belief/preference formation and diffusion in social networks, and on how this affects the group's ``epistemic potential": the ability of the agents to track the truth of the matter (with respect to some given relevant topic). While in some cases, ``wisdom of the crowds" can increase the epistemic potential, in other situations the group's dynamics leads to informational distortions: the ``madness of the crowds": cascades, `groupthink', the curse of the committee, pluralistic ignorance, group polarization, doxastic cycles etc. I explain how logic (in combination with probabilistic methods) can be used to provide some explanations for both types of situations, as well as to suggest some partial solutions to informational distortions.
>> DIEP Seminar: Michel Mandjes (U. Amsterdam)
A diffusion-based analysis of a road traffic network | 30th of September 2021
​
In this talk I will discuss an important example of complex networks, namely road traffic networks. I start by giving an overview of the existing literature, distinguishing between microscopic models (describing the stochastic evolution of the position of individual vehicles) and macroscopic models (built around deterministic continuous flows, represented as partial differential equations). Then I argue that the “optimal" model is a compromise between these: we aim at a stochastic model with enough aggregation to make sure that explicit limiting analysis can be performed. The underlying dynamics are consistent with the macroscopic fundamental diagram that describes the functional relation between the vehicle density and velocity. Discretizing space, the model can be phrased in terms of a spatial population process, thus allowing the application of a classical scaling approach. More specifically, it follows that under a diffusion scaling, the vehicle density process can be approximated by an appropriate Gaussian process. This Gaussian approximation can be used to evaluate the travel time distribution between a given origin and destination. Based on joint work with Jaap Storm (VU).
>> DIEP Seminar: Federica Russo (U. Amsterdam)
Disease Causation and Public Health Interventions: the Entanglement of Conceptual, Methodological, and Normative Questions | 16th of September 2021
​
In this talk, I discuss how concepts, methods, and values are entangled. While the argument can be applied widely across the sciences, I focus here on the sciences of health and disease, and on public health. In particular, I will show that different ways of conceptualizing (disease) causation are inherently linked to the methods (used in the health sciences) and to values (at work e.g. in public health). I will argue, on top of well-established arguments, not only that scientific methods and concepts are value-laden, but also value-promoting, and so any normative questions cannot be asked at the end or outside of the scientific process, but should be an integral part of it.
>> DIEP Seminar: Maris Ozols (U. Amsterdam)
Introduction to quantum circuits | 24th of June 2021
​
Any Boolean function can in principle be decomposed into elementary logical gates, such as AND, OR and NOT, that act only on one or two bits at a time. Similarly, any operation on a quantum computer can be broken up into elementary gates that act only on one or two qubits at a time. I will explain how this works and what consequences this has for quantum algorithm design.
>> DIEP Seminar: Casper van Elteren (U. Amsterdam)
Through the looking glass - Information flows in complex systems | 10th of June 2021
​
Understanding dynamical systems is a fundamental problem for the 21st century.
Despite the prima facie differences and purposes of many real-world networks,
previous research shows several universal characteristics in networks properties
such as the small-world phenomenon, fat-tail degree and feedback loops. This has
lead to the common but often implicit assumption that the connectedness of a
node in the network is proportional to its dynamic importance. For example in
epidemic research, high degree nodes or "super-spreaders" are associated to
dominant epidemic risk and therefore deserve special attention. Yet prior
research shows that the shared universality in network characteristics is not
shared in the dynamic or functional properties of many real-world systems.
In this talk I will explore the relation between local interactions and
macroscopic properties of a system through the lens of statistical physics and
information theory. In particular, I will show novel methods on determining the
so-called driver node in complex systems, and how tipping point can be studied
from an information theoretical perspective.
>> DIEP Seminar: Velimir Ilić (Mathematical Institute of the Serbian Academy of Sciences and Arts)
An overview and characterization of generalized information measures | 3rd of June 2021
​
The aim of this talk is to present a comprehensive classification of the main entropic forms introduced in the last fifty years within statistical physics and information theory and to review the fundamental questions about the meaning of information. I will particularly focus on axiomatic approaches to the characterization of various generalizations of the Shannon entropy, such as the Rényi, the Tsallis, the Sharma-Mittal, and the Sharma-Mittal-Taneja entropies, as well as the more general classes of pseudo-additive entropies with a well-defined mathematical and information theoretic structure. Finally, I will point out possible applications of these measures in communication theory, statistical inference and complex systems modelling.
>> DIEP Seminar: Olivier Roy (U. Bayreuth)
Deliberation, Coherent Aggregation, and Anchoring | 27th of May 2021
​
In this talk we will present a number of results stemming from a computational model of collective attitude formation through a combination of group deliberation and aggregation. In this model the participants repeatedly exchange and update their preferences over small sets of alternatives, until they reach a stable preference profile. When they do so the collective attitude is computed by pairwise majority voting. The model shows, on the one hand, that rational preference change can fill an existing gap in known mechanisms purported to explain how deliberation can help avoiding incoherent group preferences. On the other hand, the model also reveals that when the participants are sufficiently biased towards their own opinion, deliberation can actually create incoherent group rankings, against the received view. The model suggests furthermore that rational deliberation can exhibit high levels of path dependencies or "anchoring", where the group opinion is strongly dependent on the order in which the participants contribute to the discussion. We will finish by discussing possible trade-offs between such positive and negative features of group deliberation.
>> DIEP Seminar: Sven Banisch (Max Planck Institute Leipzig)
Social Feedback Theory: Modeling collective opinion phenomena by learning from the feedback of others | 20th of May 2021
​
Humans are sensitive to social approval and disapproval. The feedback that others provide on our expressions of opinion is an important driver for adaptation and change. Social feedback theory provides a framework for modeling collective opinion processes based on these principles. The theory departs from previous models by differentiating an externally expressed opinion from an internal evaluation of it. Opinion dynamics is conceived as repeated games that agents play within their social network and to which they adapt by reward-based learning. Within this setting, game theoretic notions of equilibrium can be used to characterize structural conditions for qualitatively different regimes of collective opinion expression. In this talk, I aim for a broad perspective and discuss two models addressing emergent phenomena such as polarization and collective silence.
>> DIEP Seminar: Alan Kirman (Aix-Marseille University)
Crises in a complex world | 6th of May 2021
In the summer of 2019 three well known economists , Suresh Naidu, Dani Rodrik and Gabriel Zucman published an article in the Boston Review, entitled, “Economics After Neoliberalism” in which they argued that contemporary economics is finally breaking free from its market fetishism, offering plenty of tools we can use to make society more inclusive.
In response a number of us published, in the same journal, the following plea to go further.
“Our backgrounds are in economics, political science, psychology, anthropology, physics, computer science, evolutionary theory, and complex systems theory. To us, the phenomenon called “the economy” is a highly complex, multilevel system that encompasses human biology, human behavior, group behavior, institutions, technologies, and culture, all mutually entangled in networks of nonlinear, dynamic feedback. Each of these levels in the system is subject to learning, adaptation, evolutionary, and coevolutionary processes, which means that the system is constantly changing, self-creating, and never at rest. These dynamics in turn create system-level emergent behaviors, including economic growth, inequality, and financial booms and busts. The whole system, in turn, is deeply embedded in the physical processes of our planet.
This transdisciplinary perspective, sometimes referred to as “complexity economics,” differs in a number of significant ways from the traditional perspective of economics.”
Eric Beinhocker, W. Brian Arthur, Robert Axtell, Jenna Bednar, Jean-Philippe Bouchaud, David Colander, Molly Crockett, J. Doyne Farmer, Ricardo Hausmann, Cars Hommes, Alan Kirman, Scott Page, and David Sloan Wilson.
The purpose of my presentation is to show some of the ways in which complexity economics which views aggregate behaviour as emerging from the interaction between individuals and institutions can help us to understand the evolution of our socio-economic system. This will help us to break out of the restrictive framework in which crises are the result of exogenous shocks and lead us to think of crises as endogenous and arising from the way in which the system self organises.
>> DIEP Seminar: Jasper van Wezel (U. Amsterdam)
The non-Hermitian "split skin effect" | 15th of April 2021
From atomic chains, to lattices of cold atoms and metamaterials, non-reciprocal and non-conservative systems hosting waves can exhibit a dramatic phenomenon, known as the non-Hermitian skin effect in which all bulk modes are forced to one side of a finite system. Here, we demonstrate a driven mechanical chain that hosts a "split skin effect", in which an extensive fraction of the bulk modes localises on the side of the system opposite to the usual bulk mode localisation, and opposite to the driving in the chain.
​
This system realises a specific instance of a broad class of non-Hermitian, non-reciprocal systems in both classical and quantum mechanics, whose dynamics is governed by Toeplitz matrices. We present a theoretical analysis highlighting how both normal and split skin effect phases may arise in these systems, and how the localisation length of the various skin effect modes depends on the properties of the underlying Toeplitz matrix. Although our results clearly show the skin effect is not by itself topological in nature, we suggest an interpretation of the skin modes as topological edge modes of a hypothetical higher-dimensional system.
>> DIEP Seminar: Joris Mooij (U. Amsterdam)
How to learn causal relations from data | 22nd of April 2021
​
Many questions in science, policy making and everyday life are of a causal nature: how would a change of A affect B? Causal inference, a branch of statistics and machine learning, studies how cause-effect relationships can be discovered from data and how these can be used for making predictions in situations where a system has been perturbed by an external intervention. In this talk, I will introduce the basics of two, apparently quite different, approaches to causal discovery. I will discuss how both approaches can be elegantly combined in Joint Causal Inference (JCI), a novel constraint-based approach to causal discovery from multiple data sets. This approach leads to a significant increase in the accuracy and identifiability of the predicted causal relations. One of the remaining big challenges is how to scale up the current algorithms such that large-scale causal discovery becomes feasible.
>> DIEP Seminar: Mazi Jalaal (U. Amsterdam)
Light Production in a Unicellular Organism | 8th of April 2021
Bioluminescence (emission of light from living organisms) is a common form of communication in the ocean and land. It has evolved over forty times in history and can be found in multiple biological kingdoms like bacteria, fungi, and protozoa. Bioluminescence has been of interest to humankind for thousands of years and has been a source of commentary since ancient times, from Aristotle and Pliny the Elder to Shakespeare, Boyle and Darwin. While the internal biochemistry of light production by many organisms is well established, the manner by which fluid shear or mechanical forces trigger bioluminescence is still poorly understood. We will briefly review the history of the science of bioluminescence and then present our recent work on the bioluminescence of a single-celled organism, where we aim to understand the response (light production) to mechanical stimulation. We find a "viscoelastic" response in which light intensity depends on both the amplitude and rate of cell deformation, consistent with the action of stretch-activated ion channels. We also show how such a biological system can be modelled with a simple set of linear ordinary differential equations.
>> DIEP Seminar: Florian Wagener (U. Amsterdam)
How good risk management can cause financial crises | 1st of April 2021
Risk management is as old as finance. It is based on the observation that total risk can be reduced by spreading it out over market participants. However, new financial instruments are regularly at the root of global financial crises. We propose a mechanism how the availability of more financial instruments may destabilise markets when traders have heterogeneous expectations and adapt their behaviour according to performance-based reinforcement learning.
>> DIEP Seminar: Sebastian De Haro (U. Amsterdam)
Formulating Emergence in the Physical Sciences -- a Philosopher's Perspective | 25th of March 2021
An important problem in the philosophy of emergence is the different uses that different authors make of the word ‘emergence’, and of the distinctions that they draw between different kinds of emergence. In this talk, I will review recent proposals, especially by Butterfield, to define emergence as novelty of behaviour relative to an appropriate comparison class, and to clarify the relation between emergence, reduction, and supervenience. Then I will present my own proposal for how to best define emergence in the physical sciences.
>> DIEP Seminar: Mohsen Sadeghi (Freie Universität Berlin)
Large-scale dynamics of biomembranes and membrane-associated proteins | 18th of March 2021
Lipid bilayer membranes are self-assembled structures with the thickness of a few nanometers that can form an assortment of geometries of several micrometers in size, vital to the function of living cells. The so-called peripheral proteins, that can bind to the surface of membranes, are responsible for shaping and remodeling biomembranes. The necessary cooperative action of a multitude of curvature-inducing proteins leads to the emergence of the rich membrane geometries observed in living cells. In order to model this highly dynamic system close to its native spatiotemporal scales, a mesoscopic model that can accurately mimic membrane mechanics and solvent hydrodynamics is needed. In this talk, I present our coarse-grained dynamic membrane model [1], and the corresponding approach to hydrodynamics that leads to realistic membrane kinetics [2]. I will talk about the entropic membrane-mediated interactions and investigate the kinetics, stationary distributions, and the free energy landscape governing the formation and break-up of protein clusters on the surface of the membrane.
​
[1] M. Sadeghi, T. R. Weikl, and F. Noé. Particle-based membrane model for mesoscopic simulation of cellular dynamics. J. Chem. Phys., 148(4):044901, 2018.
​
[2] Mohsen Sadeghi and Frank Noé. Large-scale simulation of biomembranes incorporating realistic kinetics into coarse-grained models. Nat. Commun., 11(1):2951, 2020.
>> DIEP Seminar: Clélia de Mulatier (U. Amsterdam)
Beyond pairwise model for binary data: the search for simple spin models| 11th of March 2021
Finding the model that best captures the patterns hidden within noisy data is a central problem in science. To address this issue, information theory and Bayesian statistics provide two comparable rigorous methods to select the best of potential explanations for data. The selected model is the one that achieves the optimal balance between goodness-of-fit and simplicity. Yet in practice, the computational cost associated with fitting each of the many potential models and the difficulty of evaluating model complexity make it challenging to search for “the” best model. Besides, with a finite amount of data an important limitation comes from the large degeneracy of models that perform nearly optimally.
In this talk I will discuss these issues in the context of binary data, where pairwise spin models (Ising model) are widely used. To understand the features of simple models, we will study the information theoretic complexity of spin models with interactions of arbitrary order, which form a complete family of candidate models for binary data. We will highlight the existence of transformations between models with interactions of different orders that preserve model complexity and see that, contrary to common intuition, pairwise models are not necessarily the simplest spin models.
We will finally discuss the development of new complementary methods of model selection for binary data that take into account high order interactions. In particular, we will discuss the use of minimally complex models for which all quantities of interest – the model complexity, the maximum likelihood, the evidence, and the Fisher information matrix – can be computed easily. This approach contrasts with the statistical inference of pairwise models for which maximum likelihood estimates are already computationally challenging. We will illustrate these techniques on several datasets.
>> DIEP Seminar: Wout Merbis (DIEP)
Exact epidemic models form a tensor product formulation | 24th of February 2021
A method for computing exact transition rate matrices for many well-known models of epidemic spreading on networks is presented. The state of the population is described as a tensor product of N individual probability vector spaces, with dimension equal to the number of compartments of the epidemiological model d. The transition rate matrix for the d^N-dimensional Markov chain is obtained by taking suitable linear combinations of tensor products of d-dimensional matrices. The resulting transition rate matrix is a sum over bilocal linear operators, which gives insight into the microscopic dynamics of the system. We show how the exact transition rate matrix for the susceptible-infected (SI) model can be used to find analytic solutions for SI outbreaks on finite trees and the cycle graph. We comment on possible applications of this formulation to the study of stochastic systems with many interacting constituents, such as the epidemic spreading process and other models of information spread on networks.
>> DIEP Seminar: Vítor Vasconcelos (U. Amsterdam)
Coordination in polarised societies| 18th of February 2021
Polarisation on various issues has increased in many western democracies since the 1980s. Beliefs about or observations of the behaviours and opinions of others drive individuals’ actions. These multiple social dynamics can support cooperative equilibria in the absence of enforcement by formal institutions, but they can also maintain harmful beliefs and behaviours. Through modelling and experiments, we explore the effects of polarisation on the likelihood that a society will coordinate on welfare-improving actions in a context where collective benefits are acquired only if enough individuals contribute—i.e., a coordination game. The talk will start with an analysis of competing complex-contagion processes and their role in generating different patterns in the distribution and segregation of ideas or opinions. Then, we will look into how heterogeneity of these opinions impacts collective action and the role of partial—and biased—information about others in improving the chances of collective success. Finally, we will show how different types of biases, not just the ones introduced by limited information but those intrinsic to human psychology, lead to suboptimal deadlocks.
>> DIEP Seminar: Clara Stegehuis (U. Twente)
Network structure and the spread of epidemics | 11th of February 2021
Many real-world networks contain groups of densely connected nodes, also called communities. We use random graph models to show that these community structures strongly influence the behavior of epidemic processes on networks: community structures can both enforce as well as inhibit epidemic processes. Our models further show that the exact internal structures of communities barely influence the behavior of percolation processes across networks. We then investigate how the final size of an epidemic is influenced by contact tracing and quarantining. We show that the effectiveness of such tracing processes strongly depends on the network structure. In contrast to previous findings, the tracing procedure is not necessarily more effective on networks with heterogeneous degrees. We also show that network clustering influences the effectiveness of the tracing process in a non-trivial way: depending on the infectiousness parameter, contact tracing on clustered networks may either be more, or less efficient than on networks without clustering.
_edited.jpg)
>> DIEP Seminar: Janusz Meylahn (DIEP)
Algorithmic collusion using Q-learning | 4th of February 2021
Algorithmic pricing is becoming more and more integrated into the marketplace. The danger of this, according to some economists and lawyers, is that the algorithms may learn to collude spontaneously. This would take the form of the algorithms charging higher prices than would be competitive. In the article we will discuss this week, Calvano et al. conduct simulated experiments with a Q-learning algorithm. They show that the use of the algorithm by two firms in a duopoly leads to supra-competitive prices without the algorithm being explicitly programmed to do so. Surprisingly, the algorithm seems to respond to an exogenous "defection" with a "punishment" followed by a return to the supra-competitive price. A strategy of this kind (Reward and Punishment Scheme) is thought to be a crucial ingredient for stable collusion. The result that such a strategy spontaneously emerges when using Q-learning has received sufficient attention to warrant a follow-up publication in Science.
>> DIEP Seminar: Soroush Rafiee Rad (DIEP)
On Correlated Information | 28th of January 2021
We discuss a paper by Alexandru Baltag and Sonja Smets on correlated knowledge: https://www.researchgate.net/publication/226943513_Correlated_Knowledge_An_Epistemic-Logic_View_on_Quantum_Entanglement
​
In this paper they model (classical and quantum) complex systems, and give a logical analysis of classical and quantum correlations using tools developed in the study of epistemic logics. They propose a logical system for reasoning about the information carried by a complex system consisting of different parts, and investigate the relationship between the information available in such a system as a whole and the information carried by each of its parts. In particular, their analysis distinguishes distributed information, that comes from pooling together all the information that can be observed in each separate part of the system, from correlated information, that is obtained by joint observations of the parts. This correlated information is only obtainable when the individual parts are combined and observed as a complex system, which allows information exchange between these parts. Similarly, for a set of individual agents with private information, correlated knowledge only emerges when they come together as a group, allowing for cooperation between them and the information dynamics that ensue. This is an instance of an emergent phenomena in information dynamics and epistemic logic and an example of logical analysis of such phenomena. This analysis elucidates the difference between classical and quantum information and provides an informational-logical characterization of 'quantum entanglement’.
>> DIEP Seminar: Tom Lancaster (Durham U.)
Topological order in quantum and classical physics | 21st of January 2021
The fractional quantum Hall (FQH) fluid has been advanced by several authors as a candidate for demonstrating strong emergence, most notably through its hosting particles with fractional electronic charges. The properties of the FHQ fluid can be traced back to topological order: a non-symmetry breaking ordering of electrons that relies on whole-system interactions. Although this exotic state of affairs might seem a quirk of quantum mechanics, this is not the case. In this talk I shall discuss the possibility of emergence in both the FQH fluid and, in an example drawn from the classical world of soft-matter physics, in a system of entangled polymer rings.
.jpeg)
>> DIEP Seminar: Mauricio Del Razo (DIEP)
Molecular kinetics as hybrid switching diffusions: a general framework for MSM/RD simulation | 14th of January 2021
A novel approach to simulate simple protein-ligand systems at large time- and length-scales is to couple Markov state models (MSMs) of molecular kinetics with particle-based reaction-diffusion (PBRD) simulations; this approach is named MSM/RD. Current formulations of MSM/RD lack an underlying mathematical framework; they are limited to protein-ligand systems, where the ligand orientation and conformation switching are not taken into account; and they lack multi-particle extensions. In this work, we develop a general MSM/RD framework by coarse-graining molecular dynamics into hybrid switching diffusion processes, a class of stochastic process that integrates continuous dynamics and discrete events into the same process. This framework overcomes all the previous difficulties. We further implement and verify it for two toy models of protein-protein interaction and one multiparticle implementation to model the formation of pentameric ring molecules.
