Visiting Speaker
Benjamin Fish
University of Illinois at Chicago
Inferring network structure from data
Thu
Feb 22, 2018
Download TALK slides
3:00 pm
177 Huntington Ave
11th floor
ADD to calendar

When is it possible to infer network structure from data?  In the first part of the talk, we investigate how to reconstruct social networks from voting data. In particular, given a voting model that considers social network structure, we aim to find the network that best explains the agents’ votes.  For two models of voting, we give algorithms and lower bounds, characterizing cases where network recovery is possible and where it is computationally difficult.  Despite the similarity of the two models, we show that their respective network recovery problems substantially differ in complexity and produce very different networks, and then show the results on real data.  In the second part of the talk, we discuss how to learn the time scale of a dynamic network.  As a pre-processing step before data analysis, typically you need to choose a bin size, where all edges occurring within a bin are aggregated together to form a sequence of graphs.  In previous work, finding a bin size is often accomplished with an unsupervised heuristic, which does not take into account that each task an analyst wants to perform on the data requires a different bin size.  Given this, we introduce new windowing algorithms that automatically adapt to the task the analyst wants to perform by treating windowing as a hyperparameter for the task.  We show this approach outperforms previous approaches on real data.

about the speaker
Benjamin Fish is a PhD candidate at the University of Illinois at Chicago and a member of the Mathematical Computer Science group. His research focuses on the theory and practice of machine learning, using data that can actually be collected in real-world settings. This includes learning over social networks, adaptive data analysis, and algorithmic fairness. He earned a B.A. from Pomona College in Mathematics and Computer Science. He has received the NSF EAPSI Fellowship and has been a visiting researcher at both the University of Melbourne and the University of Utah. For more details, visit http://ben.fish/

No upcoming events

Thank you! You have been added to our email list.

Oops! Something went wrong while submitting the form