University of Minnesota
University Relations
myU OneStop

Go to unit's home.

Home | Seminars and Symposia | Past seminars/symposia: Thursday, October 2, 2014

DTC Seminar Series

The Dynamics of Social Learning and Sensing


Vikram Krishnamurthy
Department of Electrical and Computer Engineering
University of British Columbia

Thursday, October 2, 2014
2:20 p.m. reception
2:45 p.m. seminar

401/402 Walter Library

Statistical inference via social networks presents several unique challenges from a statistical signal processing point of view. Agents (viewed as social sensors) interact with and influence other agents. Also agents usually reveal decisions and not private observations. There is strong motivation to construct models that capture the interacting dynamics of multiple agents in social networks, together with algorithms that can be used to estimate events of interest. This talk is comprised of three highly stylized examples involving social learning, asymptotic approximations to large random graphs and game-theoretic learning:

The first part of the talk deals with the dynamics of Bayesian social learning together with methods for mitigating herding and data incest.

The second part of the talk deals with models for the diffusion of information in large scale social networks.These use the mean-field dynamics of random graphs to model the interaction of agents that influence each other. Examples include the spread of information on social media, localization and tracking using social networks.

In the third part of the talk, we consider how multiple agents interacting over a social network can coordinate their actions to achieve a rational outcome. We will introduce game-theoretic learning algorithms that provably converge to a set of correlated equilibria — which is a generalization of Nash equilibria — and an ideal analysis tool for modeling individuals that interact and learn.

Details of the talk can be found in the following monograph and papers:

  1. V. Krishnamurthy, O Namvar and M. Hamdi, Signal Processing of Social Networks: Interactive Sensing and Learning, Foundations and Trends in Signal Processing, 2014.
  2. V. Krishnamurthy and H.V. Poor, A Tutorial on Interactive Sensing in Social Networks, IEEE Transactions Computational Social Systems, Vol.1, No.1, 2014.
  3. V. Krishnamurthy, Quickest Detection POMDPs with Social Learning: Interaction between local and global decision makers, IEEE Transactions on Information Theory, Vol.58, No.8, pp.5563--5587, 2012.
  4. V. Krishnamurthy, Bayesian Sequential Detection with Nonlinear Penalty - A POMDP Lattice Programming Approach,IEEE Transactions on Information Theory, Vol.57, No.10, pp.7096–7124, 2011.
  5. O. Namvar, V. Krishnamurthy and G. Yin, Distributed Tracking of Correlated Equilibria in Noncooperative Games, IEEE Transactions Automatic Control, Vol.58, No.10, pp.2435–2450, 2013.


Vikram Krishnamurthy holds the Canada Research Chair in signal processing at the Department of Electrical and Computer Engineering, University of British Columbia, Canada. His current research interests include social networks, computational game theory and stochastic control. He has served as Distinguished lecturer for the IEEE signal processing society and as Editor in Chief of IEEE Journal Selected Topics in Signal Processing. In 2013 he was awarded an honorary doctorate from KTH (Royal Institute of Technology), Sweden.