Intermittent updates
On October 9th, we are lucky to have Pantelis Pipergias Analytis visiting the group. Pantelis recently moved as an assistant professor at the Danish Institute of Advanced Studies (D-IAS) at the University of Southern Denmark.
Before moving to Denmark, he spent the past two years as a postdoctoral researcher at the Computer and Information Science department at Cornell University. Pantelis got his PhD from the Max Planck Institute for Human Development in Berlin.
Pantelis will give a talk based on his recent Nature Human Behavior paper Social learning strategies for matters of taste
Title: Social learning strategies for matters of taste
Abstract: Most choices people make are about ‘matters of taste’, on which there is no universal, objective truth. Nevertheless, people can learn from the experiences of individuals with similar tastes who have already evaluated the available options—a poten- tial harnessed by recommender systems. We mapped recommender system algorithms to models of human judgement and decision-making about ‘matters of fact’ and recast the latter as social learning strategies for matters of taste. Using computer simulations on a large-scale, empirical dataset, we studied how people could leverage the experiences of others to make better decisions. Our simulations showed that experienced individuals can benefit from relying mostly on the opinions of seemingly similar people; by contrast, inexperienced individuals cannot reliably estimate similarity and are better off picking the main- stream option despite differences in taste. Crucially, the level of experience beyond which people should switch to similarity- heavy strategies varies substantially across individuals and depends on how mainstream (or alternative) an individual’s tastes are and the level of dispersion in taste similarity with the other people in the group.
Our old friend Piotr, current postdoc at Northeastern, and graduate from the group is visiting from his new home beyond the Atlantic. This coming Thursday, Piotr will give a short about his most recent work. Details below.
Abstract: Ranked lists of persons and items are a core part of the user experience in many online services, such as search, social media feeds, hiring, and dating sites. Studies have shown disparate amount of attention received by high rank results, potentially leading to loss of opportunity and access to resources among the lower ranked items. In this short talk I will give an overview of the work on individual and group fairness in ranked lists and focus on our work in progress: a novel metric for investigating group unfairness in ranked lists. Our approach relies on estimating the amount of attention given to members of a protected group and comparing it to that group’s representation in a defined population. It offers two major developments compared to the state of the art. First, rather than assuming a logarithmic loss in importance as a function of the rank, we allow for attention distributions that are specific to the audited service and the habits of its users. For example, more items are consumed in a single viewing of a social media feed than as a result of a single query in a web search engine. Second, we allow non-binary protected attributes (gender, race, etc.), both to better reflect the way individuals identify, but also to enable measurements on aggregates of multiple search runs, rather than separately for each result list.We investigate the properties of the metric and compare them to the behavior of other established approaches using synthetic ranked lists. Finally, we showcase the metric through a simulated audit of a number of hiring and dating services.
Later this month we will have legendary researcher Bernardo Huberman visiting. And we’re lucky enough to have him giving a talk on one of the most exciting new developments in Network Science: Applying AI to networking problems.
Bernardo has been a central player throughout the rise of network theory (and mentor for field notables, such as Lada Adamic and Jure Leskovec), but that’s just a fraction of what he’s accomplished. If you care about anything related to information sciences, this is a talk you cannot miss.
Bernardo is a Fellow and vice president of the Core Innovation Team at CableLabs. He is also a Consulting Professor in the Department of Applied Physics and the Symbolic System Program at Stanford University. Previously he was Senior Fellow and Senior Vice President at Hewlett Packard Enterprise Company, and Director of the Mechanisms and Design Lab at Hewlett Packard Labs.
Title: Artificial Intelligence and the Network
Abstract: Artificial Intelligence is the attempt to make computers emulate human cognition and thought processes. It has existed for a long time and has sprouted a number of subfields, from semantic networks and common sense reasoning to robotics, logic programming and machine learning. In spite of the glacial rate of progress in AI, one subfield, machine learning, has recently taken off like wildfire. What powers this incredible growth is the availability of fast processors that have made possible computations than seemed hard to achieve a few years ago. As a result, we now have powerful systems that can easily recognize myriad images and spoken languages. This talk will describe some of the great successes of machine learning, their limitations, and their application to networking problems which pervade modern communications. I will also present a form of artificial intelligence that is distributed in nature and that mimics the ability of groups of people and social insects to solve extremely hard problems.
We’re lucky to have Max Schich visiting DTU tomorrow. Max is an associate professor for arts and technology at The University of Texas at Dallas and a founding member of the Edith O’Donnell Institute of Art History. His work converges hermeneutics, information visualization, computer science, and physics to understand art, history, and culture. Schich is the first author of “A Network Framework of Cultural History” (Science magazine, 2014) and a lead co-author of the animation “Charting Culture” (Nature video, 2014). He is an editorial advisor at Leonardo Journal, an editorial board member at Palgrave Communications (NPG), and the Journal for Digital Art History. He publishes in multiple disciplines and speaks to translate his ideas to diverse audiences across academia and industry. His work received global press coverage in 28 languages.
Details
ABSTRACT: History has no periodic table of elements and no theory of temporal structure, as George Kubler pointed out in 1962, yet, as he also points out, things occupy time in a bounded number of ways. The obvious question still is: Can we capture the shape of time? – Tackling this challenge, this talk looks at historical time systematically, dealing with more or less exponential growth, the archaeological paradox, global and meso-level patterns, cycles, periodicity, condensation, and a bouquet of oddities.
Here’s a cool video about some of Max’s recent work
Sometime last year I became an adjunct professor at University of Copenhagen’s Department of Sociology. And just to be clear: I’m still primarily the Technical University of Denmark. The adjunct position is more of a way of signaling that I work closely with social science researchers (e.g. through my associate director position at SODAS).
Anyway, the important thing here is that I’m finally giving my inaugural lecture. The lecture is a fun chance for me to reflect on what’s happened up to now. My goal is to make the lecture be fun, entertaining, and personal (in a way that I hope will shed light on the mechanics of the scientific process). I hope you’ll come and see it.
Here are the details:
The amazing Villum Foundation (who first funded the SensibleDTU project back in 2012) wanted to showcase some of the work coming out of their Young Investigator Program. As part of that, they have produced a beautiful video about my research (and also a bit about who I am).
It’s directed by Ole Stenum who shot something like 5 hours of conversation and elegantly captured the essence the whole thing in a 3 minute film. It was fun to try to be part of the whole process.
This Wednesday we will have Benjamin Maier, a PhD student (physics) from Dirk Brockmann’s group speaking at DTU. Ben is based at the Robert Koch-Institut but also affiliated with Humboldt University Berlin’s physics department and IRI Life Sciences. His interest lies in identifying the underlying processes of human contact facilitating the spread of diseases. In particular he is investigating the influence of structural properties of human contact networks or human transport networks, both of static and dynamic nature.
Abstract: Studying the dynamics of face-to-face interaction networks is essential for a better understanding of contact mediated processes, contagion processes, and disease spreading. In many studies regarding social systems, networks are reconstructed using time averages or integrated networks, in which links reflect an interaction likelihood, although frequently this measure is not well defined but serves as a qualitative feature from which network properties are computed. During the last years a significant effort was made to resolve this issue by developing algorithms to analyse dynamic processes on the actual time-dependent contact patterns of social systems. However, there is still a lack of simple dynamic network models generating temporal networks of typical behaviour observed in real systems.
We introduce a class of minimal dynamic network models that naturally yield group formation and are easy to control. In those models, randomly chosen individual nodes cut their existing links, connect to a target node and establish links to the target’s neighbors. We discuss a variety of properties of those models and show how to use it for comparison of epidemic processes on real-world data.
Next week, I’ll be giving a talk about SensibleDTU at DTU. I hope the text below explains everything.
Abstract: We’ve actually been doing a lot of interesting work on the Sensible DTU dataset over the past year or so. (SensibleDTU is the project where we collected + dynamic multilayer network and behavioral data from 1000 smartphones) . This talk covers highlights and goes in depth with the most exciting projects. And you have a chance to ask questions: Maybe there’s something in there for you to test your own algorithm on. I’m also considering delivering some deep and personal revelations.
Notice: The talk will be filmed as part of a movie created by the Villum Foundation, who funded a large chunk of Sensible DTU. And – if you agree – you might end up in the film.
For the concluding speaker in our September-talks series, I am happy to announce that Hartmut Lentz will talk about his work on spread of infectious disease on temporal networks. Hartmut works at the “Institut für Epidemiologie” at the Friedrich Loeffler Institute. He is a fantastic speaker and an authority on temporal networks. Full details below.
Abstract: Many networks are treated as static objects, although they are in fact strongly time-dependent. This can have a dramatic impact on the possible spreading patterns for infectious diseases.
A static (aggregated) trade network is constructed as follows: if two nodes are connected directly to each other in a time-dependent network, the same connection is present in the static network. A fundamental difference between the static and the time dependent view however, is the consideration of paths, i.e. indirect connections over more than one edge. Concerning paths, the causality of the edges used plays an essential role. In an aggregated network, paths can seem causal, although they do not follow a time-respecting sequence of edges in the real system. This leads to a systematic overestimation of outbreak sizes, if time-dependent networks are treated as static.
We introduce a new method, which allows for the computation of the total causal path structure of a temporal network (represented by its accessibility graph) using the adjacency matrices of its snapshots. In addition, information about the timescales required for path traversal can be derived from the step-by-step derivation of the accessibility graph of the network. This procedure directly yields the distribution of shortest path durations in a temporal network. In addition, we define the new measure causal fidelity that compares the number of paths in a temporal network with its aggregated counterpart. This measure allows a quantitative assessment of how well a temporal network can be approximated by a static aggregated one.
The methods presented here require only basic knowledge linear algebra and can be implemented efficiently. Their capability is demonstrated for three examples: networks of social contacts, livestock trade, and sexual contacts.
