Learning from Collective Behavior

Inspired by longstanding lines of research in sociology and related fields, and by more recent largepopulation human subject experiments on the Internet and the Web, we initiate a study of the computational issues in learning to model collective behavior from observed data. We define formal models for efficient learning in such settings, and provide both general theory and specific learning algorithms for these models.

• Type: Conference
• Year: 2008
• Where: COLT
• Authors: Michael Kearns, Jennifer Wortman
• Keywords: COLT 2008, Efficient Learning, Longstanding Lines, Machine Learning, Specific Learning Algorithms