>> DIEP seminars and talks - archives 2023
Below you can find more information on the DIEP seminars of 2023.
>> DIEP Seminar: Eric Dignum (University of Amsterdam)
Computational modelling of school choice and school segregation: from theoretical to data-driven agent-based models | 11am, 14th of December 2023
​
Many educational systems still consist of substantial levels of school segregation along various lines, such as race, ethnicity, household income levels, parental educational attainment and ability. This means that, globally, pupils with similar characteristics cluster together in the same schools. This is widely acknowledged to reproduce or even exacerbate inequalities and result in unequal outcomes. Hence, even despite decades of research and policies aimed to counteract school segregation, it is still a persistent societal problem. Existing literature shows that factors affecting school segregation and individual components in the system of school choice interact with each other. These interactions are reasoned to be an important mechanism through which the levels of school segregation emerge on the macro-level, but commonly used qualitative and quantitative analysis methodologies (e.g. discrete choice models, interviews, surveys) often ignore these dependencies between the components and their consequences (e.g. feedback loops, non-linearity). Using methodologies that ignore these interactions could be an explanation why school segregation is still plaguing our educational systems. However, tools from complex systems such as Agent-Based Models (ABMs), can potentially be a complementary methodology that can explicitly account for such interactions and could therefore complement our understanding of the effect of them on school segregation. In this talk, two examples of ABMs are presented which illustrate that a complexity perspective on school choice dynamics can meaningfully contribute to the field of school choice and school segregation. Firstly, using a highly stylised ABM, an alternative explanation of why schools are often more segregated relative to neighbourhoods is given. Households do not have to be less tolerant for school compositions compared to neighbourhood compositions. Instead, this ABM demonstrates that asymmetric preferences are not a requirement for excess school segregation, but that residential segregation combined with distance preferences and interacting components can also play a key role. However, these highly stylised models have limited applicability to reality and actual policy scenarios. Therefore, we also present ongoing work on a methodology to empirically calibrate large-scale ABMs on empirical data in the context of school choice. We show that this methodology is able to retrieve the (artificial) ground truth within reasonable accuracy. Additionally, (computational) challenges and open questions are discussed, focusing on the usage of household-level register data.
>> DIEP Seminar: Christian List (LMU Munich)
Do group agents have free will | 11am, 30th of November 2023
​
It is common to ascribe agency to some organized collectives, such as corporations, courts, and states, and to treat them as loci of responsibility, over and above their individual members. But since responsibility is often assumed to require free will, should we also think that group agents have freewill? Surprisingly, the literature contains very few in-depth discussions of this question (a notable exception is Hess 2014). I will argue that group agents can have free will not only in a relatively undemanding compatibilist sense, but also in a recognizably libertarian sense, which includes the possibility of doing otherwise. In developing this account of corporate free will, I will bring together recent work on group agency and recent work on free will.
>> DIEP Seminar: Renee Hoekzema (VU Amsterdam)
Multiscale methods for gene selection in single cell transcriptomics data | 11am, 23rd of November 2023
​
Single cell transcriptomics is a revolutionary technique in biology that allows for the measurement of gene expression levels in many individual cells simultaneously. Analysis of these large datasets reveals complex variation in expression patterns between cells. Current methods for analysis assume that cell types are discrete. However, in practice there is also continuous variation between cells: subtypes of subtypes, differentiation pathways, responses to environment or treatment, et cetera. The complexity found in modern single cell transcriptomics datasets calls for intricate methods to biologically interpret both discrete clusters as well as continuous variations. We propose topologically-inspired data analysis methods that identify coherent gene expression patterns on multiple scales, considering discrete and continuous patterns on equal footing. As well as finding new biologically meaningful genes, the methodology allows one to visualise and explore the space of gene expression patterns in the dataset.
>> DIEP Seminar: Meike Wortel (University of Amsterdam)
Eco-evolutionary mechanisms to explain species and strain diversity | 11am, 16th of November 2023
​
I will discuss a recent paper https://onlinelibrary.wiley.com/doi/full/10.1111/jeb.14158 where I investigated whether fluctuating nutrients can (partly) explain diversity observed in microbial communities, such as oceans or the human gut. When nutrient levels fluctuate over time, one possibly relevant mechanism is coexistence between specialists on low and specialists on high nutrient levels. The relevance of this process is supported by the observations of coexistence in the laboratory, and by simple models, which show that negative frequency dependence of two such specialists can stabilize coexistence. However, as microbial populations are often large and fast growing, they evolve rapidly. I will discuss how we determine what happens when species can evolve; whether evolutionary branching can create diversity or whether evolution will destabilize coexistences. We derive an analytical expression of the invasion fitness in fluctuating environments and use adaptive dynamics techniques to find that evolutionarily stable coexistence requires a special type of trade-off between growth at low and high nutrients. I will also discuss current work on a different mechanism: When one phenotype degrades a toxin which allo another phenotype to grow (cross-protection) and ideas how to generalize results from both of these studies to study the interaction of more mechanisms in larger networks.
>> DIEP Seminar: Christian Hamster (University of Amsterdam)
Journal club on modelling evolutionary dynamics | 11am, 9th of November 2023
​
Real world ecosystems can be very complex, with many species coexisting that are functionally very similar. For example, there are many different plankton species in the oceans competing for a small number of resources. Mathematical modelling, however, indicates that only a few species should survive, the 'most fit' species. This discrepancy is known as the paradox of the plankton. Our approach starts with a simple resource-consumer model, and at a random timepoint, we introduce a new species that is a small random perturbation of an already existing species. This allows us to numerically mimic evolution and build complex ecosystems. This approach raises a lot of questions, both mathematically and biologically. From a mathematical perspective, can we understand and describe the numerical solutions? And from a biological perspective, what constitutes a species in our system? Is every perturbation a new species, or just genetic variation within a species and should we look at clusters of species? These and many other questions we can discuss during the meeting.
_edited.jpg)
>> DIEP Seminar: Han van der Maas (University of Amsterdam)
Sociophysics: an overview and new ideas | 11am, 2nd of November 2023
​
To develop a mathematical and computational toolbox for studying the interdependencies between polarization, segregation, and inequality, an overview of modeling in sociophysics is needed. In this talk I will try to provide such an overview. I will also present some new research directions in the field of polarization.
>> DIEP Seminar: Liang Li (Max Planck Institute of Animal Behavior)
Schooling fish: from biology to robotics and back | 11am, 26th of October 2023
​
With over half a billion years behind them, fish have evolved to swim with remarkable efficiency, agility, and stealth in their three-dimensional aquatic world. Given this, it's natural that engineers often look to fish for inspiration when developing underwater propulsion systems. Over the years, roboticists have been inspired by these biological marvels to design fish-like robots that mimic real fish in terms of morphology, locomotion, and movement. Interestingly, the trend has recently shifted from merely drawing inspiration from biology to using robotics as a tool for better understanding biological processes. In this talk, I will first discuss our approach to designing and controlling these robotic fish, rooted in the concept of bio-inspiration. I will then provide examples of how we employ both real and virtual robots to investigate the mechanisms of collective behaviour in schooling fish. To conclude, I'll offer a glimpse into my current and future endeavors in the realms of robotics and biology.
>> DIEP Seminar: Alexandru Baltag & Sonja Smets
(University of Amsterdam)
Logic Meets Wigner’s Friend(s): the epistemology of quantum observers | 11am, 19th of October 2023
​
This presentation is about Wigner's Friend thought-experiment [1], and its more recent variations [2] and extensions such as the Frauchiger-Renner (FR) Paradox [3], that have recently reignited the debates in the foundations of quantum theory. Such thought experiments seem to indicate that, if quantum theory is assumed to be universally valid (and hence can be applied to multi-partite systems that may include classical observers), then different agents are rationally entitled to ascribe different (mutually inconsistent) states to the same system, and as a result they cannot share their information in a consistent manner. More precisely, the result in [3] is stated in the format of a no-go theorem. To analyze this problem, we focus on a few questions: what is the correct epistemic interpretation of the multiplicity of state assignments in these scenarios?; under which conditions can one include classical observers into the quantum state descriptions, in a way that is still compatible with Quantum Mechanics?; under which conditions can one system be admitted as an additional ‘observer’ from the perspective of another (background) observer?; when can the standard axioms of multi-agent Epistemic Logic (that allow for “knowledge transfer” between agents) be applied to quantum-physical observers? After discussing some of the various answers to these questions proposed in the literature, we propose a new such answer, sketch a particular formal implementation of it, and apply it to obtain a principled solution to Wigner Friend-type paradoxes. The presentation is based on recent joint work [4].
[1] E.P. Wigner, Remarks on the mind-body question, in I.J. Good, The Scientist Speculates, London Heinemann,1961.
[2] D. Deutsch, Quantum theory as a universal physical theory, International Journal of Theoretical Physics, 24, I, 1985.
[3] D. Frauchiger and R. Renner, Quantum theory cannot consistently describe the use of itself, Nature Communications, 9(1):3711. Preprint, arXiv:1604.07422, 2018.
[4] A. Baltag and S. Smets, Logic meets Wigner’s Friend (and their Friends), preprint, arXiv:2307.01713 [quant-ph], July 2023.
>> DIEP Seminar: Charlotte Hemelrijk (University of Groningen)
Self-organized motion and collective escape in schools of fish and flocks of birds | 11am, 12th of October 2023
​
It is mysterious how schools of fish and flocks of birds, such as the huge flocks of starlings, coordinate their members particularly when the shape and density of the flock change miraculously, for instance during collective escape from a predator by collective turning, flock-splitting, flash expansion and the so-called wave of agitation. These fascinating questions need to be solved by combining computational models with empirical data. Empirical data have been collected on many aspects, such as the shapes of the flock, the degree of motion of individuals within a flock and the collective patterns of evasion of a predator, for different species of birds when they are chased by a robotic bird. In the present talk, we will use computational models based on self-organization, such as StarDisplay and others, because flocks and schools in these models resemble empirical data in many respects. We will show what causes the shape of a fish school to be on average oblong and that of huge flocks of starlings to be continuously changing; whether or not starlings keep close to a familiar neighbour and what may be underlying in schools and flocks the internal motion, flock splitting and the wave agitation.
>> DIEP Seminar: Alexandre Genin (University of Utrecht)
Spatial self-organization and its consequences in simple and complex ecosystems | 11am, 5th of October 2023
​
Because species interact with each other, the spatial organisation of ecosystems is seldom random, but most often self-organized into specific spatial patterns. A seminal example is the case of positive interactions between plants producing emergent patches, bands, or fractal patterns in a landscape, with important consequences on its resilience to perturbations. In this seminar, we will first focus on simple, but widespread examples of such ecological interactions leading to spatial-self organisation. We will show how the non-random, emergent spatial structure can be quantified to help us understand and predict ecosystem collapse, showcasing a recent application to the coral reefs of Rapa Nui (Easter Island, Chile). We will then discuss how these relatively simple principles can be extended to species-rich systems, to understand the link between ecological interactions and spatial self-organisation, this time in more complex settings.
>> DIEP Seminar: Conor Finn (Max Planck Institute for Mathematics in the Sciences)
Pointwise information decomposition for complex systems | 11am, 28th of September 2023
​
The aim of information decomposition is to provide a mathematical framework that partitions the total information provided by a set of source variables about a target variable into i) the unique information that is provided by each individual source variable, ii) the shared information that is redundantly provided by two or more source variables, and iii) the synergistic information that is only attainable from simultaneous knowledge of two or more source variables. Such a decomposition has many potential applications in the sciences: for instance, quantifying the synergistic interaction between multiple incoming neural stimuli that are fused to create some output signal, or for determining the extent to which a particular phenotypic trait depends uniquely on each individual source gene, redundantly on two or more genes or is determined by some synergistic combination of several genes. Most approaches to information decomposition focus on decomposition the average information provided the variables involved. In this talk, I will discuss why this is unsatisfactory, especially when in comes to analysing complex systems. I will provide a brief overview of the pointwise perspective on information theory and then discuss the challenges associated with determining a pointwise information decomposition. I will then discuss the pointwise partial information decomposition that we introduced before closing by demonstrating how this theory can be used to quantify emergent intrinsic computation in canonical complex systems, namely elementary cellular automata.
>> DIEP Seminar: Daniel Miranda (Federal University of Pernambuco)
Phenomenological renormalization group analysis of cortical spiking data | 10am, 21st of September 2023
​
The critical brain hypothesis has emerged in the last decades as a fruitful theoretical framework for understanding collective neuronal phenomena. Lending support to the idea that the brain operates near a phase transition, Beggs and Plenz were the first to report experimentally recorded neuronal avalanches, whose distributions coincide with the mean-field directed percolation (DP) universality class, which comprises a variety of models in which a phase transition occurs between an absorbing (silent) and an active phase. However, this hypothesis is highly debated, as neuronal avalanches analyses and other common statistical mechanics tools may struggle with challenges ubiquitous in living systems, such as subsampling, long range correlations and the absence of an explicit model for the complete neuronal dynamics. In this context, Meshulam et al. recently proposed a phenomenological renormalization group (PRG) method to deal with neural networks typical long range interactions with a model independent analysis. The procedure consists of repeatedly manipulating the data, obtaining an increasingly coarse-grained description of the activity after each iteration. Under a critical regime, non-trivial correlations and scale-free behavior should be unveiled as we simplify our description. This can be inferred from a series of statistical features of the data, which lead us to different scaling relations. Here, we apply this phenomenological renormalization group (PRG) in different experimental setups. Additionally, we investigate how the scaling exponents found via PRG behave as we parse our data by its coefficient of variation (CV); this measurement has appeared in recent literature as a means of tracking different cortical states through spiking variability.
>> DIEP Seminar: Ro Jefferson (Utrecht University)
Physics ∩ Machine Learning | 11am, 14th of September 2023
​
Machine learning has become both powerful and ubiquitous, but remains a black box whose internal workings are still largely unclear. In this talk, I will discuss some interesting connections between ideas in physics (in particular QFT and the renormalization group) and deep neural networks in particular, which collectively motivate a physics-based approach towards a theory of deep learning.
>> DIEP Seminar: Daniele Marinazzo (University of Ghent)
(Higher-order) informational interactions: ideas, implementations, and applications in neurosciences and behavioral sciences” | 11am, 7th of September 2023
​
Systems composed of many units, whose behavior goes beyond the sum of the individual behaviors of the singles, are ubiquitous. Examples relevant to what we do are the brain, the body as a whole, and the social systems we live in. When it comes to analyzing collective behavior we are often stuck with pairwise dependencies (often correlations). In this talk, I will describe a framework rooted in information theory to mine multiplets of variables sharing common information about the variability of complex systems, and provide some examples in neuroscience, physiological, and psychometrics.
>> DIEP Seminar: Pierre Baudot (Median Technologies)
Information cohomology, higher-order statistical interactions, complexity and deep networks | 11am, 29th June 2023
​
Information theory, probability and statistical dependencies, and algebraic topology provide different views of a unified theory yet currently in development, where uncertainty goes as deep as Galois's ambiguity theory, topos and motivs. I will review some foundations, that characterize uniquely entropy as the first group of cohomology, on random variable complexes and probability laws. This framework allows to retrieve most of the usual information functions, like KL divergence, cross entropy, and was extended by Juan Pablo Vigneaux to Tsallis entropies, differential entropy. Multivariate interaction/Mutual information (I_k and J_k) appear as coboundaries, and their negative minima, also called synergy, corresponds to homotopical link configurations, which at the image of Borromean links, illustrate what purely collective interactions or emergence can be. Those functions refine and characterize statistical independence in the multivariate case, in the sens that (X1,...,Xn) are independent iff all the I_k=0 (with 1<k<n+1, whereas for Total correlations G_k, it is sufficient that G_n=0), generalizing correlation coefficient. Concerning data analysis, restricting to the simplicial random variable structure sub-case, the application of the formalism to genetic transcription or to some classical benchmark dataset using open access infotopo library, unravels that higher statistical interactions are nonetheless omnipresent but also constitutive of biologically relevant assemblies. On the side of deep networks, information cohomology provides a topological and combinatorial formalization of deep networks' supervised and unsupervised learning, where the depth of the layers is the simplicial dimension, derivation-propagation is forward (co-homological). Recently, Leon Lang could generalize higher-order mutual informations notably to Tsallis entropy and Kolmogorov complexity, and Tom Mainiero further established a general associated index, or “Euler characteristic”, given by the Tsallis mutual informations on a weighted simplicial complex whose topology retains information about the correlations between various subsystems. Tsallis mutual informations hence open a totally new and unexplored territory for higher order interactions studies, both theoretically and in application to data-natural sciences.
_edited.jpg)
>> DIEP Seminar: Abel Jansma (University of Edinburg)
The information theory of higher-order interactions | 11am, 22nd June 2023
​
Information-theoretic quantities reveal dependencies among variables in the structure of joint, marginal, and conditional entropies, but leave some fundamentally different systems indistinguishable. Furthermore, there is no consensus on how to construct and interpret a higher-order generalisation of mutual information (MI). In this talk, I will show that a recently proposed model-free definition of higher-order interactions amongst binary variables (MFIs), like mutual information, is a Möbius inversion on a Boolean algebra, but of surprisal instead of entropy. This gives an information-theoretic interpretation to the MFIs, and by extension to Ising interactions. We will study the dual objects to MI and MFIs on the order-reversed lattice, and find that dual MI corresponds to conditional mutual information, while dual interactions (outeractions) are interactions with respect to a different background state. Unlike mutual information, in- and outeractions uniquely identify all six 2-input logic gates, the dy- and triadic distributions, and different causal dynamics that are identical in terms of their Shannon-information content.
>> DIEP Seminar: Mike Lees (UvA)
Measuring, Modelling and Simulating Crowd Dynamics: Mobility to Epidemics | 11am, 15th June 2023
​
In this talk I will present the challenges of understanding and modelling the multi-disciplinary problem of human mobility and crowd dynamics. I’ll highlight our attempts to conduct empirical experiments of human crowds and demonstrate the technological and technical challenges that this presents. I’ll show the classical approaches used by computational scientists when modelling crowds and the challenges of connecting measurement to models. I’ll showcase two ongoing projects where we attempt to measure and model crowd dynamics to understand the spread of infectious disease. Firstly, the Kumbh Mela Experiment where we measure and model pilgrims in Ujjain, India, to estimate the spread of Tuberculosis. Secondly, A project in the Johan Cruyff arena where we use Wi-Fi data to try and connect movement human movement ecology (e.g., Levy Flight Dynamics) to the exposure and spread of infectious diseases.
>> DIEP Seminar: Jay Armas (DIEP)
Risk aversion promotes cooperation | 11am, 8th June 2023
​
Cooperation is at the heart of many phenomena in living and complex systems, including multicellularity, eusociality in insect societies, human communities and financial markets. Many mechanisms that lead to the emergence of cooperation in small groups have been proposed and studied in depth in the past decades. Yet, little is known about the existence of potential mechanisms that can sustain large-scale cooperation involving a very large group of individuals. I will combine chemical reaction networks with evolutionary game theory and stochastic methods to study simple models of interacting groups and show, under certain assumptions, that if individuals are risk averse, cooperation can emerge in large groups.
>> DIEP Seminar: Diego Garlaschelli (University of Leiden and the IMT School of Advanced Studies, Lucca, Italy)
Multiscale network renormalization: scale-invariance without geometry | 11am, 1st June 2023
​
Systems with lattice geometry can be renormalized exploiting their coordinates in metric space, which naturally define the coarse-grained nodes. By contrast, complex networks defy the usual techniques, due to their small-world character and lack of explicit geometric embedding. Current network renormalization approaches require strong assumptions (e.g. community structure, hyperbolicity, scale-free topology), thus remaining incompatible with generic graphs and ordinary lattices. Here we introduce a graph renormalization scheme valid for any hierarchy of coarse-grainings, thereby allowing for the definition of `block-nodes' across multiple scales. This approach reveals a necessary and specific dependence of network topology on additive hidden variables attached to nodes, plus optional dyadic factors. Renormalizable networks turn out to be consistent with a unique specification of the fitness model, while they are incompatible with preferential attachment, the configuration model or the stochastic blockmodel. These results highlight a deep conceptual distinction between scale-free and scale-invariant networks, and provide a geometry-free route to renormalization. If the hidden variables are annealed, they lead to realistic scale-free networks with density-dependent cut-off, assortatitivy and finite local clustering, even in the sparse regime and in absence of geometry. If they are quenched, they can guide the renormalization of real-world networks with node attributes and distance-dependence or communities. As an application, we derive an accurate multiscale model of the International Trade Network applicable across hierarchically nested geographic partitions.
>> DIEP Seminar: Leonardo di Gaetano (Central European University)
Percolation and Topological properties of temporal higher-order networks | 11am, 25th May 2023
​
Hypergraphs provide a more accurate representation of complex systems with non-pairwise interactions, such as social networks and cellular networks. However, analyzing and characterizing hypergraphs remains a challenge. To address this, we present a hidden variables formalism to analyze higher-order networks. We apply this framework to a higher-order activity-driven model and provide analytical expressions for the main topological properties of the time-integrated hypergraphs. Our analysis demonstrates the importance of considering higher-order interactions and shows that neglecting them can lead to underestimating the percolation threshold. Overall, our work contributes to a better understanding of the interplay between group dynamics and the unfolding dynamical processes over them in complex systems.
>> DIEP Seminar: Ana Millan Vidal (U. Granada)
The role of epidemic spreading in seizure propagation and epilepsy surgery | 11am, 11th May 2023
​
Computational models of brain dynamics can provide new insights into the prognosis of neurological disorders such as epilepsy. Epilepsy surgery is the treatment of choice for drug-resistant epilepsy patients, but up to 50% of the patients continue to have seizures one year after the resection. The propagation of seizures over the brain can be regarded as an epidemic spreading process taking place on the patient's connectome. We show in a retrospective study (N=15) that this simple dynamic -namely the Susceptible-Infected-Recover (SIR) model- is enough to reproduce the main aspects of seizure propagation as recorded via invasive electroencephalography (iEEG) [2,3]. Remarkably, the SIR model parameters that best describe the iEEG seizure patterns correspond to the critical transition between the percolating and absorbing phases of the SIR model [2,3,4], and the similarity between the iEEG and modelled seizure predicted surgical outcome (area under the curve AUC = 0.73). We validated the use of the model in the clinic with a blind, independent pseudo-prospective study (N=34) using the parameters as in the retrospective study to avoid over-fitting. As a consequence iEEG data (highly invasive and not always part of the presurgical evaluation) was not required. Using the model to find optimal resection strategies [1], we found smaller resections (AUC=0.65) for patients with good outcome, indicating intrinsic differences in the presurgical data of patients with good and bad outcome [4]. The actual resection also overlapped more with the optimal one (AUC=0.64) and had a larger effect decreasing modelled seizure propagation (AUC=0.78) for patients with good outcome [4]. Individualised computational models may inform surgical planning by suggesting optimal resection strategies and informing on the likelihood of a good outcome after a proposed resection. This is the first time that such a model is validated on a fully independent cohort without the need for iEEG recordings.
_edited.jpg)
>> DIEP Seminar: Swarnendu Banerjee (U. Utrecht)
Rethinking tipping points in spatial ecosystems | 11am, 4th May 2023
​
The theory of alternative stable states and tipping points has garnered a lot of attention in the last decades. It predicts potential critical transition from one ecosystem state to a completely different state under increasing environmental stress. However, typically ecosystem models that predict tipping do not resolve space explicitly. As ecosystems are inherently spatial, it is important to understand the effects of incorporating spatial processes in models, and how those insights translate to the real world. Moreover, spatial ecosystem structures, such as vegetation patterns, are important to predict ecosystem response in the face of environmental change. Models and observations from real savanna ecosystems and drylands have suggested that they may exhibit both tipping behavior as well as spatial pattern formations. Hence, in this talk, I will use mathematical models of humid savannas and drylands to illustrate several pattern formation phenomena that may arise when incorporating spatial dynamics in models that exhibit tipping. I will argue that such mechanisms challenge the notion of large-scale critical transitions in response to global change and reveal a more resilient nature of spatial ecosystems.
>> DIEP Seminar: Christian Hamster (Wageningen University)
Understanding Stochastic Waves in Cell Movement Models, from Gillespie Algorithms to (S)PDEs | 11am, 20th April 2023
​
Single-cell organisms are remarkably good at sensing food, especially if you consider that they lack our sensing organs and have to measure a gradient in the food supply over the length of a single cell. The precise mechanisms behind this gradient sensing are not fully understood yet, but scientists
have determined many relevant molecules that are relevant in the motion of the cell and we can see how these molecules are activated in wavelike patterns. These processes can be used to build stochastic models for cell movement, where individual molecules are modeled. These models are complex, both numerically
and analytically, so we often summarise everything into 'simpler' PDEs. In this talk, I would like to introduce (and explain) an in-between option, so-called Chemical Langevin Equations, effectively a Stochastic PDE approximation of the underlying stochastic algorithms. This approach allows us to use all the insights from the deterministic PDE, without throwing away the stochastic nature of the models.
>> DIEP Seminar: Sandro Sozzo (University of Udine)
Quantum nature of energy and entropy in cognition: Towards a Non—classical Thermodynamic Theory of Human Culture | 11am, 13th April 2023
​
Convincing evidence reveals that quantum structures model cognitive phenomena better and more efficiently than classical structures. This has led to the development of a novel research programme called quantum cognition. Inspired by a two–decade research on the field, we extend here the range of applicability of quantum cognition and prove that the notions of energy and entropy can be consistently introduced in human language and, more generally, in human culture. More explicitly, if energy is attributed to words according to their frequency of appearance in a text, then the ensuing energy levels are distributed non–classically, namely, they obey Bose–Einstein, rather than Maxwell–Boltzmann, statistics, because of the genuine quantum indistinguishability of the words that appear in the text. Secondly, the quantum entanglement due to the way meaning is carried by a text reduces the (von Neumann) entropy of the words that appear in the text, a behaviour which cannot be explained within classical (thermodynamic or information) entropy. We claim that this ‘quantum–type behaviour holds in general in human language', namely, any text is conceptually more concrete than the words composing it, which entails that the entropy of the overall text decreases as a result of composition. In addition, we provide examples taken from cognition, where quantization of energy appears in categorical perception, and from culture, where entities collaborate, thus entangle, to decrease overall entropy. We use these findings to propose the development of a non–classical thermodynamic theory for human cognition, which also covers broad parts of human culture and its artefacts, and bridges concepts with quantum physics entities.
>> DIEP Seminar: Rodrigo Cofre (Paris-Saclay University)
Novel perspectives on the structure-function dynamics of the primate and human brain under diverse consciousness states | 11am, 6th April 2023
​
The analysis of brain network dynamics provides an insightful perspective to analyze the dynamic reconfiguration in brain network structure across species to investigate its role during the loss of consciousness. Recent work has highlighted the importance of the dynamical aspect in understanding the functional relevance of alterations in this network structure to investigate how the brain supports consciousness. In this talk, my idea is to introduce the topic of research of consciousness from the STEM perspective and then fly over a set of ideas and recent results in order to discuss the problems associated with data analysis in the temporal dimension.
I will show convergent and complementary results between different methods of investigating the dynamical aspects of the structure-function relationship during the loss of consciousness in the primate and human brain. If time permits, I would like to present different ideas and datasets that we may use in collaboration to explore more in-depth the relationship between the different states of consciousness and recorded brain activity under those states.
>> DIEP Seminar: Chase Broedersz (VU Amsterdam)
The dynamics of cell migration in flat and curved geometries | 11am, 30th March 2023
​
In many biological phenomena, cells migrate through flat or curved confining environments. However, a quantitative framework to describe the stochastic dynamics of such multicellular confined cell migration remains elusive. We employ a data-driven approach to infer the dynamics of cell movement, morphology and interactions of cells confined in micropatterns. By inferring a stochastic equation of motion directly from the experimentally determined short time-scale dynamics, we show that cells exhibit intricate non-linear deterministic dynamics that adapt to the geometry of confinement. We extend this approach to interacting systems, by tracking the repeated collisions of confined pairs of cells. By inferring an interacting equation of motion for this system, we find that non-cancerous (MCF10A) cells exhibit repulsive and frictional interactions. In contrast, cancerous (MDA-MB-231) cells exhibit attraction and a novel and surprising anti-friction interaction, causing cells to accelerate upon collision. Based on the inferred interactions, we show how our framework may generalize to provide a unifying theoretical description of diverse cellular interaction behaviors. Finally, I will discuss the collective dynamics cells migrating in 3D curved confining geometries in multicellular spheroids.
>> DIEP Seminar: Ricard Solé (ICREA-Complex Systems Lab)
Emergence, tinkering and universality in evolved networks | 11am, 23rd of March 2023
​
A common trait of complex systems is that they can be represented using a network of interacting parts. In fact, the network organization (more than the parts) largely conditions most higher-level properties, which are not reducible to the properties of the individual parts. Can the topological organization of these webs provide some insight into their evolutionary origins? Both biological and artificial networks share some common architectural traits. They are often heterogeneous and sparse, and most exhibit different types of correlations, such as nestedness, modularity or hierarchical patterns. These properties have often been attributed to the selection of functionally meaningful traits. However, a proper formulation of generative network models suggests a somewhat different picture. Against the standard selection–optimization argument, some networks reveal the inevitable generation of complex patterns resulting from reuse and can be modelled using duplication–rewiring rules lacking functionality. In other examples, such as human language, information tradeoffs might be responsible for the presence of universal scaling laws. Both give rise to the observed heterogeneous, scale-free and modular architectures. Here, we examine the evidence for tinkering and universality in cellular, technological and ecological webs and its impact on shaping their architecture. We suggest that both tinkering and information constraints shape these graphs at the topological level. In biological systems, selection forces would take advantage of emergent patterns.
>> DIEP Seminar: Anshul Toshniwal (U. Amsterdam)
Opinion Dynamics in Populations of Converging and Polarizing agents | 11am, 16th March 2023
​
Opinions determine individuals' attitudes and fundamentally influence collective decisions in societies. As a result, understanding the processes leading to the dynamic formation of opinions is a key research topic across multiple disciplines, from sociology and political sciences to multi-agent systems and statistical physics. Opinion dynamics have been simulated through different computational models where agents are assumed to interact over networks and be influenced through their social ties. Often, models assume that agents with opposing viewpoints converge in opinion when interacting with each other. This is at odds with evidence showing that individuals can also become further polarized when connected with individuals having opposing viewpoints, suggesting the existence of converging individuals (that become less radicalized when interacting with opposing agents) but also polarizing individuals (that become more radicalized in such settings). In this talk I will describe opinion dynamics when converging and polarizing nodes co-exist in a population. Through simulations and dynamic systems analysis we will try to understand 1) how radicalization depends on different combinations of such type of nodes and 2) how placing polarizing/converging agents in specific network locations impacts opinion radicalization. We observe that there is an optimal fraction of polarizing nodes that minimizes radicalization. Furthermore, we observe that placing polarizing nodes on specific network positions can strongly affect radicalization: assigning high-degree nodes as polarizing results in lower radicalization as compared to random assignment. Our results indicate that considering heterogeneous agents in what concerns their reaction to opposing viewpoints is fundamental to fully grasp the role of social networks in sustaining radical opinions.
>> DIEP Seminar: Iain Couzin (Max Planck Institute of Animal Behaviour)
The Geometry of Decision-Making | 11am, 9th March 2023
​
Running, swimming, or flying through the world, animals are constantly making decisions while on the move—decisions that allow them to choose where to eat, where to hide, and with whom to associate. Despite this most studies have considered only on the outcome of, and time taken to make, decisions. Motion is, however, crucial in terms of how space is represented by organisms during spatial decision-making. Employing a range of new technologies, including automated tracking, computational reconstruction of sensory information, and immersive ‘holographic’ virtual reality (VR) for animals, experiments with fruit flies, locusts and zebrafish (representing aerial, terrestrial and aquatic locomotion, respectively), I will demonstrate that this time-varying representation results in the emergence of new and fundamental geometric principles that considerably impact decision-making. Specifically, we find that the brain spontaneously reduces multi-choice decisions into a series of abrupt (‘critical’) binary decisions in space-time, a process that repeats until only one option—the one ultimately selected by the individual—remains. Due to the critical nature of these transitions (and the corresponding increase in ‘susceptibility’) even noisy brains are extremely sensitive to very small.
>> DIEP Seminar: Vincent Buskens (Utrecht U.)
Disease avoidance may come at the cost of social cohesion: Insights from a large-scale social networking experiment | 11am, 2nd March 2023
​
It is known that people tend to limit social contact during times of increased health risks, thus leading to the disruption of social networks and changing the course of epidemics. It is, however, less known to what extent people show such avoidance reactions. To test the predictions and assumptions of an agent-based model on the feedback loop between avoidance behavior, social networks, and disease spread, we conducted a large-scale (2879 participants) incentivized experiment. The experiment rewards maintaining social relations and structures, and penalizes acquiring infections. We find that disease avoidance dominates networking decisions, despite relatively low penalties for infections; and that participants use more sophisticated strategies than expected to prevent infections, while they forget to maintain a profitable network structure. Consequently, we observe lower numbers of infections than predicted, but also deterioration of network positions. These results imply that the focus on a more obvious signal (i.e., disease avoidance) may lead to unwanted side effects (i.e., loss of social cohesion).
_edited.jpg)
>> DIEP Seminar: Fernando Rosas (Imperial College)
Formal approaches to emergence: theory, practice, and opportunities | 11am, 9th February 2023
​
Emergence is a profound subject that straddles many scientific scenarios and disciplines, including how galaxies are formed, how flocks and crowds behave, and how human experience arises from the orchestrated activity of neurons. At the same time, emergence is a highly controversial topic, surrounded by long-standing debates and disagreements on how to best understand its nature and its role within science. A way to move forward in these discussions is provided by formal approaches to quantify emergence, which give researchers new frameworks to guide discussions and advance theories, and also quantitative tools to rigorously establish conjectures about emergence and test them on data. This talk presents an overview on the theory and practice of these formal approaches to emergence, and highlights the opportunities they open for practical data analysis. We elaborate on their unifying principles and the distinctive benefits of them, and present illustrative examples of their application. We finish discussing several interpretation issues and potential misunderstandings, and presenting ideas about how to further develop these research efforts for the benefit of empirical inquiry.
--tojpeg_1495816919459_x1_edited.jpg)
>> DIEP Seminar: Fabian Greimel (U. Amsterdam)
Falling Behind: Has Rising Inequality Fueled the American Debt Boom? | 11am, 2nd February 2023
​
We evaluate the hypothesis that rising inequality was a causal source of the US household debt boom since 1980. The mechanism builds on the observation that households care about their social status. To keep up with the ever richer Joneses, the middle class substitutes status-enhancing houses for status-neutral consumption. These houses are mortgage-financed, creating a debt boom across the income distribution. Using a stylized model we show analytically that aggregate debt increases as top incomes rise. In a quantitative general equilibrium model we show that Keeping up with the Joneses and rising income inequality generate 60% of the observed boom in mortgage debt and 50% of the house price boom. Finally, we provide novel empirical evidence on the relationship between top incomes and household debt. Mortgage debt rose substantially more in US states that experienced stronger growth in top incomes. There is no such relationship between top incomes and non-mortgage debt. These findings support to the importance of the comparisons channel.
>> DIEP Seminar: Marco Javarone (Centro Ricerche Enrico Fermi in Rome and University College London)
Evolutionary Game Theory beyond Cooperation | 11am, 26th January 2023
​
Evolutionary Game Theory (EGT) allows for facing the challenge of cooperation. For instance, EGT studies strategies and mechanisms able to trigger the emergence of cooperation in populations whose interactions rely on dilemma games, e.g. the Public Goods Game, whose Nash equilibrium is typically defection. Yet, although cooperation is a fundamental open challenge in science, this talk aims to show cross-disciplinary applications and results beyond this problem. To this end, after a brief general overview of this field, I will present some works that use EGT for analysing complex phenomena in ecology, epidemiology and blockchain dynamics.
>> DIEP Seminar: Anton Souslov (U. Bath)
Topological fibre optics | 11am, 19th January 2023
​
A challenge in photonics is to create a scalable platform in which topologically protected light can be transmitted over large distances. I will talk about the design, modeling, and fabrication of photonic crystal fibre (PCF) characterised by topological invariants [1]. The fibre is made using a stack-and-draw technique in which glass capillaries are stacked, molten, and drawn to desired size. Light propagates in glass cores, whose normal modes are analogous to atomic orbitals. Topological invariants emerge in the band structure of many coupled cores inside a periodic array, analogous to an atomic crystal. We directly measure the bulk winding-number invariant and image the associated boundary modes predicted to exist by bulk-boundary correspondence. The mechanical flexibility of fiber allows us to reversibly reconfigure the topological state. As the fiber is bent, we find that the edge states first lose their localization and then become relocalized because of disorder. We envision fiber as a scalable platform to explore and exploit topological effects in photonic networks.
