Invited Speakers

Anxo Sánchez

Bio: PhD in Theoretical Physics (with distinction) from Universidad Complutense de Madrid, Spain, 1991. Fulbright postdoctoral fellow at Los Alamos National Laboratory, USA, 1993 - 1994. Currently, full professor of Applied Mathematics at Universidad Carlos III de Madrid, and founder of the research group GISC in 1996. Associated researcher of the Institute for Biocomputation and Physics of Complex Systems (BIFI), Universidad de Zaragoza.

His research deals mostly with the applications of physics of complex systems to social and biological sciences; he has contributed to the advancement of different fields, from economics to condensed matter physics through ecology and theoretical computer science. During his career he has obtained relevant results for the different communities interested. Examples include the largest experiments ever carried out on cooperation on networks, the mathematical demonstration of the existence of Dunbar circles in personal relationships and of a counterpart of the circles in nonhuman primates, or the experimental measurement of the evolution of social norms.

Website: https://anxosanchez.eu/

Talk title: There is more to signed networks than social balance (by Miguel A. González-Casado, Andreia Sofia Teixeira and Angel Sánchez)

Abstract: This talk concerns itself with signed networks, their dynamics and the mechanisms behind them from a layman’s view point. We begin by introducing a rich, longitudinal dataset on friendships and enmities at a high school which we study from an empirical viewpoint. We subsequently discuss how to model it by using signed networks, and in so doing we identify a number of mechanisms and phenomena that play a role, going beyond the traditional focus on social balance as the key driver of network evolution. As a consequence of the interaction of these mechanisms, we then show that the social network is very accurately in equilibrium over the years, irrespective of kids leaving or entering the school, and that this equilibrium state is not balanced. We thus conclude that approaching a problem with a signed network model may benefit from a naive viewpoint that is open to including many features beyond social balance.

Anna Gallo

Bio: She is currently a PhD student at the IMT School For Advanced Studies Lucca (Italy) working under the supervision of Professors Tiziano Squartini and Diego Garlaschelli. She holds a Bachelor’s degree (2019) and a Master’s degree (2021) in Mathematics at the University of Florence (2021). Her Master’s thesis focuses on the analysis of the metastable behavior of the Potts model on a square lattice. She has a record of publications on both mathematics and physics journals.

Her research interests concern probability theory, statistical physics of networks, and opinion dynamics. Her PhD focuses on the study of signed networks, the goal being that reformulating the most popular social theories within a statistical framework at both microscopic and mesoscopic levels.

Website: Google Scholar

Talk title: Testing structural balance theories in undirected and directed signed networks

Abstract: Most of the analyses concerning signed networks have focused on the balance theory [1] which formalizes the principles “the friend of my friend is my friend”; and “the enemy of my enemy is my friend”. To make balance theory testable, one needs 1) a proper representation of social networks, 2) a definition of `frustration' and 3) a set of null models to quantify the statistical significance of the latter. While the first two ingredients have been already explored comprehensively, the third one is way less developed, since the existing null models typically do not account for the intrinsically different tendencies of individual actors to establish positive and/or negative interactions. To reduce this gap, here we extend the framework of Exponential Random Graphs to binary, undirected and directed, signed networks with both global and local constraints. Moreover, we define two variants per benchmark: one where the topology is kept fixed and one where it is left to vary along with the edge signs. The level of frustration characterizing real-world, undirected, signed networks is commonly quantified via the abundance of triadic motifs. Our analysis in [2] shows that (homogeneous) null models with global constraints tend to favour the weak version of the balance theory, according to which only the triangle with one, negative link should be under-represented in real-world networks; on the other hand, (heterogeneous) null models with local constraints tend to favour the strong version of the balance theory, according to which the triangle whose links are all negative should be under-represented as well. As a comparison, biological networks display almost inverted patterns, confirming that structural balance inherently distinguishes social networks from other types of signed networks. When dealing with directed, signed networks, instead, the vast majority of the existing works has simply ignored the directionality of the edges. According to our results, however, analyzing the balance theory on directed, signed networks is not so straightforward [3] as frustration becomes a multi-faceted concept, admitting different definitions at different scales. If we limit ourselves to consider cycles of length two, frustration is related to reciprocity, i.e. the tendency of edges to admit the presence of partners pointing in the opposite direction. As the reciprocity of signed networks is still poorly understood, we adopt a principled approach for its study, defining quantities and introducing models to consistently capture empirical patterns of the kind. We find that the (directed extension of the) balance theory is not capable of providing a consistent explanation of the patterns characterizing real-world, directed, signed networks. Although part of the ambiguities can

be solved by adopting a coarser definition of balance, our results call for a theory accounting for the directionality of edges in a coherent manner. In any case, the evidence that the empirical, signed networks can be highly reciprocated leads us to recommend to explicitly account for the role played by bidirectional dyads in determining frustration at higher levels (e.g. the triadic one).

[1] Heider. The Journal of Psychology (1946).

[2] Gallo, Garlaschelli, Lambiotte, Saracco, Squartini. Communications Physics (2024).

[3] Gallo, Saracco, Lambiotte, Garlaschelli, Squartini. under submission at PRX, arXiv: 2407.08697 (2024).

George T. Cantwell

Bio: George Cantwell is an Assistant Professor in the Department of Engineering at the University of Cambridge. He specializes in networks, statistics, algorithms, physics, and behavior. George holds a Ph.D. in Physics from the University of Michigan and completed a postdoctoral fellowship at the Santa Fe Institute. His research focuses on understanding complex systems and their dynamics.

Website: https://www.george-cantwell.com/

Talk title: Defining and detecting balance

Abstract: In social networks we have intuitive assumptions about how patterns of friends and antagonism should be distributed. For example, it seems more probable that a friend-of-a-friend would also be a friend of yours, rather than an enemy. Whether this is actually true, of course, is an empirical question. Answering this question requires us to be clear about what we actually mean.

Strong balance and weak balance are two competing frameworks. But the world is messy, and neither notion holds exactly in real data. I will outline some subtleties in applying these notions to data and an inference-based solution for detecting and classifying patterns of balance.

Chiara Boldrini

Bio: Chiara Boldrini is a Senior Researcher at IIT-CNR. Her research interests include human behavioral/cognitive models for the analysis and design of online social networks, human-centric decentralized AI, causal learning in pervasive systems, and social network structures in the Metaverse. She is the IIT-CNR co-PI for the National Extended Partnership in Artificial Intelligence FAIR, H2020 SoBigData++, and H2020 HumaneE-AI-Net projects, and has been involved in several EC projects since FP7. She currently holds the position of Editor-in-Chief for Special Issues at Elsevier Computer Communications and is a member of the Editorial Board of Elsevier Pervasive and Mobile Computing. She served as TPC chair of IEEE PerCom’24 and has been on the organizing committee of several IEEE and ACM conferences/workshops, including IEEE PerCom and ACM MobiHoc. Recently, she has served in the TPC of AAAI ICWSM (as a senior PC member), The Web Conference, ECML-PKDD, WSDM, among several others.

Website: https://www.iit.cnr.it/en/chiara.boldrini/

Talk title: Keep your friends close, and your enemies closer: signed ego networks on Twitter/X

Abstract: Imagine a map of your closest social relationships, where closeness isn’t just about how often you interact, but also about the sentiment you share. This talk explores the Ego Network Model (ENM), a powerful tool for understanding social structures, with a new twist: Signed Ego Networks (SENs). SENs use sentiment analysis to reveal the positive or negative nature of our social connections. Using Twitter data from a wide range of users, we embark on a journey to understand the unique dynamics of SENs. In this talk, we will discuss how to construct signed ego networks, explore their main properties and limitations, and highlight potential applications and new research directions.