Dr. Vittoria Colizza is Director of Research at Inserm (French National Institute of Health and Medical Research) & Sorbonne University, Faculté de Médecine, Paris, France. With a PhD in Statistical and Biological Physics (SISSA, Trieste, Italy, 2004), she spent 3 years in the US at the Indiana University School of Informatics (Bloomington, IN), as post-doc and Visiting Assistant Professor, before moving to Europe and joining ISI Foundation (Turin, Italy), after being awarded a Starting Independent Career Grant in Life Sciences by the European Research Council in 2007. In 2011, Colizza joined Inserm in Paris where she leads the EPIcx lab (Epidemics in complex environments, www.epicx-lab.com) within the Pierre Louis Institute of Epidemiology and Public Health.
Her work focuses on real episodes of human and animal epidemics (e.g. 2009 H1N1 pandemic influenza, MERS-CoV epidemic, Ebola virus disease epidemic, rabies, bovine brucellosis) to gather context epidemic awareness and provide risk assessment analyses for preparedness, mitigation, and control. Her research also includes more theoretical approaches for the modeling of small- to large-scale diffusion events where contacts between hosts and their mobility are explicitly considered from data (face-to-face interactions, contact matrices, commuting, air travel, migrations, trade movements, call detail records, etc.).
Colizza received several awards, including the Young Talent Award by the Italian Ministry of Youth in 2010, the Prix Louis-Daniel Beauperthuy 2012 by the French Academy of Sciences, the Young Scientist Award for Socio-Econophysics in 2013, the Telethon Farmindustria Award in 2017, the Erdős–Rényi Prize by the Network Science Society in 2017. She also served as Young Advisor to the Vice President of the European Commission Mrs. Neelie Kroes for the Digital Agenda for Europe, and was member of the I7 Innovators’ Strategic Advisory Board on People-Centered Innovation for the Italian Government delegation for G7 in 2017.
Our understanding of communicable diseases prevention and control is rooted in the theory of host population transmission dynamics. The network of host-to-host contacts along which transmission can occur drives the epidemiology of communicable diseases, determining how quickly they spread and who gets infected. A large body of epidemiological, mathematical and computational studies has provided a number of insights into the understanding of the process and the identification of efficient control strategies. The explosion of time resolved contact data has however opened the stage to new challenges. What are the structural and temporal aspects, and possibly their non-trivial interplay, that are critical for disease spread ? To answer this question, I will introduce the infection propagator approach, a theoretical analytical framework for the assessment of the degree of vulnerability of a host population to disease epidemics, once we account for the time variation of its contact pattern. By reinterpreting the tensor formalism of multilayer networks, this approach allows the analytical computation of the epidemic threshold for an arbitrary time-varying network of host contacts, i.e. the critical pathogen transmissibility above which large-scale propagation occurs. I will apply this framework to a set of empirical time-varying contact networks and show how it can be used to test different intervention strategies for infection prevention and control in realistic settings.
AAlto University, Finland
Aristides Gionis is a professor in the department of Computer Science in Aalto University. His previous appointments include being a visiting professor in the University of Rome and a senior research scientist in Yahoo! Research. He is currently serving as an action editor in the Data Management and Knowledge Discovery journal (DMKD), an associate editor in the ACM Transactions on Knowledge Discovery from Data (TKDD), and an associate editor in the ACM Transactions on the Web (TWEB). He has contributed in several areas of data science, such as algorithmic data analysis, web mining, social-media analysis, data clustering, and privacy-preserving data mining. His current research is funded by the Academy of Finland (projects Nestor, Agra, AIDA) and the European Commission (project SoBigData).
Online social media are a major venue of public discourse today, hosting the opinions of hundreds of millions of individuals. Social media are often credited for providing a technological means to break information barriers and promote diversity and democracy. In practice, however, the opposite effect is often observed: users tend to favor content that agrees with their existing world-view, get less exposure to conflicting viewpoints, and eventually create "echo chambers" and increased polarization. Arguably, without any kind of moderation, current social-media platforms gravitate towards a state in which net-citizens are constantly reinforcing their existing opinions.
In this talk we present our ongoing line of work on analyzing and moderating online social discussions. We first consider the questions of detecting controversy using network structure and content. We then address the problem of designing algorithms to break filter bubbles, reduce polarization, and increase diversity. We discuss a number of different strategies such as user and content recommendation, as well as approaches based on information cascades.
Oxford University, UK
Prof Heather Harrington is a Royal Society University Research Fellow and Associate Professor in the Mathematical Institute at the University of Oxford. She is Co-Director of the Centre for Topological Data Analysis. Her research focuses on the problem of reconciling models and data by extracting information about the structure of models and the shape of data. To develop these methods, Prof Harrington integrates techniques from a variety of disciplines such as computational algebraic geometry and topology, statistics, optimisation, network theory, linear algebra, and dynamical systems. Based on this research, she was recently awarded a London Mathematical Society Whitehead Prize.
Persistent homology (PH) is a technique in topological data analysis that allows one to examine features in data across multiple scales in a robust and mathematically principled manner, and it is being applied to an increasingly diverse set of applications. We investigate applications of PH to dynamic biological networks with concrete examples from contagions, neuroscience, and blood vessels.
Technical University of Denmark, Denmark
Sune Lehmann is an associate professor at the Technical University of Denmark, an adjunct (full) professor at University of Copenhagen's Department of Sociology, and an adjunct associate professor at the Niels Bohr Institute (Department of Physics, University of Copenhagen). Sune is the associate director of the Center for Social Data Science at University of Copenhagen. Sune's work focuses on the dynamics of complex networks as well as processes unfolding on such evolving networks. He is the author of multiple highly cited papers and his work has received world-wide press coverage.
In other to understand the multi-layered and dynamic social interactions within a large social system, I equipped 1000 freshmen students at the Technical University of Denmark with top-of-the-line smartphones running custom software designed to collect interactions mediated through face-to-face meetings (proximity estimated via Bluetooth), telecommunication (phone-calls, text messages), and online social networks (Facebook friendships and interactions). The phones also collected geo-locations, wifi-signals, and a number of other data channels; participants also answered paneled questionnaires regarding personality, study habits, and health-related behavior. The data collection lasted 2.5 years. Through this rich dataset, we have learned about much more than social networks. In my talk, I will discuss key findings from this study, with an emphasis on communities in dynamic networks and recent results on human mobility.
City College of New York, USA
Hernan research focuses on the theoretical understanding of Complex Systems from a Statistical Physics viewpoint. He is working towards the development of new emergent laws for complex systems, ranging from brain networks to biological networks and social systems. Treating these complex systems from a unified theoretical approach, he uses
concepts from statistical mechanics, network and optimization theory, machine learning, and big-data science to advance new views on complex systems and networks.
Identifying essential nodes in complex networks is a central problem for biological systems to social systems. We treat this problem in three paradigmatic cases: the brain, ecosystems and social networks. Mathematically, we find the set of influential nodes by optimizing the damage to the giant connected component with systematic inactivation of nodes. We then apply network theory and pharmacogenetic interventions in a rat brain to predict and target essential nodes responsible for global integration in a model of learning and memory. We find that the integration of the brain network is mediated by a set of weak nodes through optimization of influence in optimal percolation. Pharmacogenetic inhibitions confirm the theoretical predictions. We discuss the relevance of these influencers to ecological systems dominated by abrupt first order tipping points as well as connectomes with regularities.
Universitat Politècnica de Catalunya, Spain
Romualdo Pastor-Satorras (Barcelona, Spain, 1967) received a PhD in Condensed Matter Physics from the Universitat de Barcelona in 1995. He spent four years as postdoctoral researcher at the Massachusetts Institute of Technology (1996-1998) and The Abdus Salam International Centre for Theoretical Physics, ICTP (1998-2000). At present, he is Associate Professor of Applied Physics at the Universitat Politècnica de Catalunya since 2006. He has been visiting scientist at, among others, Yale University (USA), the University of Notre Dame (USA), the Kavli Institute for Theoretical Physics (USA), the Helsinky University of Technology TKK (Finland), Indiana University (USA) and the Institute for Scientific Interchange (ISI) Foundation (Italy). He has been awarded twice with the national “ICREA Academia Prize” by the Government of Catalonia. He has published more than 100 publications in peer-reviewed journals in the field of statistical physics, and is author of the book “Evolution and Structure of the Internet” (Cambridge University Press, 2007), together with Professor Alessandro Vespignani.
Collective motion in animals is able to produce such stunning patterns as flocks of birds turning in unison or shoals of fish splitting and reforming while outmaneuvering a predator. The study of these phenomena is mainly based in simple models, a paradigmatic example being the one proposed by Vicsek and collaborators in the 90s. The main assumption of this and similar models is that individuals tend to orient their velocity parallel to the average velocity of their local neighborhood. The Vicsek model predicts a phase transition between an ordered phase, with individuals travelling in a common direction, and a disordered one, that has been recently the subject of a large interest in the statistical mechanics community. Here we will consider variations of the Vicsek model incorporating social interactions between individuals, represented in terms of a complex social network. The main result of the numerical study of this model is the observation that the heterogeneity of the social network can increase the resilience of the ordered state, making it immune to external perturbations. A related scalar version of the Vicsek model in networks allows for a mathematical treatment that lends support to the numerical observations, and allows for further generalizations.
RWTH Aachen University, Germany
Markus Strohmaier is the Professor for Methods and Theories of Computational Social Sciences and Humanities at RWTH Aachen University (Germany), and the Scientific Coordinator for Digital Behavioral Data at GESIS - Leibniz Institute for the Social Sciences. Previously, he was a Post-Doc at the University of Toronto (Canada), an Assistant Professor at Graz University of Technology (Austria), a visiting scientist at (XEROX) Parc (USA), a Visiting Assistant Professor at Stanford University (USA) and the founder and scientific director of the department for Computational Social Science at GESIS (Germany). He is interested in applying and developing computational techniques to research challenges on the intersection between computer science and the social sciences / humanities.
Homophily can put minority groups in social networks at a disadvantage by restricting their ability to establish links with people from a majority group. This can limit the overall visibility of minorities in the network, and create biases. In this talk, I will show how the visibility of minority groups in social networks is a function of (i) their relative group size and (ii) the presence or absence of homophilic behavior. In addition, the results show that perception biases can emerge in social networks with high homophily or high heterophily and unequal group sizes, and that these effects are highly related to the asymmetric nature of homophily in networks. This work presents a foundation for assessing the visibility of minority groups and corresponding perception biases in social networks in which homophilic or heterophilic behaviour is present.
UMass Amherst, USA
Professor Towsley's research spans a wide range of activities from stochastic analyses of queueing models of computer and telecommunications to the design and conduct of measurement studies. He has performed some of the pioneering work on the exact and approximate analyses of parallel/distributed applications and architectures. More recently, he pioneered the area of network tomography and the use of fluid models for large networks. He has published extensively, with over 150 articles in leading journals. PhD Computer Science, University of Texas (1975), BA Physics, University of Texas (1971). Professor Towsley first joined the faculty at the University of Massachusetts in the Department of Electrical and Computer Engineering in 1976 and moved to the College of Information and Computer Sciences in 1986. He was named University Distinguished Professor of Computer Science in 1998. Professsor Towsley was a Visiting Scientist at the IBM T.J. Watson Research Center, (1982-83, 2003), INRIA and AT&T Labs - Research (1996-97), and Cambridge Microsoft Research Lab (2004); a Visiting Professor at the Laboratoire MASI, Paris, (1989-90). Professor Towsley has been an editor of the IEEE Transactions on Communications, IEEE/ACM Transactions on Networking, and Journal of Dynamic Discrete Event Systems. He is currently on the Editorial boards of Networks and Performance Evaluation. He was a Program Co-chair of the joint ACM SIGMETRICS and PERFORMANCE '92 conference. He is a two-time recipient of the Best Paper Award of the ACM Sigmetrics Conference. He is a Fellow of the IEEE and of the ACM. He is also a member of ORSA and is active in the IFIP Working Groups 6.3 on Performance Modeling of Networks and 7.3 on Performance Modeling. Towsley is the recipient of one of the IEEE's most prestigious honors, the 2007 IEEE Koji Kobayashi Computers and Communications Award. He also received a UMass Amherst Distinguished Faculty Lecturer award in 2002 and a UMass Amherst College of Natural Sciences and Mathematics Faculty Research Award in 2003.
Complex networks that occur in nature, such as those from biochemistry, neurobiology, and engineering, often exhibit simple, network structural properties, or “motifs.” Network motifs refer to recurring, significant patterns of interaction between sets of nodes and represent basic building blocks of graphs. Motifs in social networks exhibit spatial patterns and temporal patterns that vary according to the type of network. This talk reports on these variations across several network types and identify several common substructures. Reciprocity of directed ties occurs much more frequently than expected by chance in all networks. Similarly, we find that completely connected triads and tetrads (i.e., four-node sub-graphs) occur more often than expected, highlighting the tendency of actors to form clusters of ties. We also identify motifs that suggest patterns of hierarchy. Motifs are also useful for the purpose of sub-graph classification. We demonstrate their value in identifying the type of network that a sub-graph belongs to.
We also consider the challenge of characterizing motifs in large graphs, and show how carefully designed sampling algorithms can accurately characterize them using a small number of samples.
Last, we close with open problems regarding motifs whose solution can lead to better understanding of social networks and analytical tools for characterizing them.