PhD Proposal: Drawing Inferences about Large Populations: Game-Theoretic and Machine Learning Perspectives

Talk
Soham De
Time: 
05.31.2017 10:00 to 11:30
Location: 

AVW 3258

In recent years, there has been an increasing connection between the social and behavioral sciences and computer science. Social science theory on human interaction has helped enrich models in the computer science community, while techniques from computer science have helped social scientists use large-scale data and agent-based models to gain more insights about human behaviors and tendencies. In this thesis proposal, this connection is further explored by:

Using computational techniques in game theory to build models of human behavior in different societies drawing from theoretical research in cross-cultural psychology. This will provide insights into the substantial societal differences that exist in how individuals interact and influence each other and how populations evolve over time, with important implications for designing policies and avoiding conflicts around the world.
Developing efficient numerical optimization techniques to make machine learning algorithms faster and more accessible to non-experts. This will enable these algorithms to process large-scale datasets more efficiently, and help non-experts to effectively use these algorithms for their research to draw inferences from massive datasets such as those generated on online social media websites.

Examining Committee:

Chairs: Dr. Dana Nau, Dr. Tom Goldstein

Dept rep: Dr. David Jacobs

Member: Dr. Michele Gelfand