Prof. Sarit Kraus

My main research interest is in Artificial Intelligence and in particular Multi-agent systems.


My main research interest in Artificial Intelligence and in particular Multi-agent systems. I focus on understanding how we can best create intelligent agents that can interact proficiently with people. We study both cooperative and conflicting scenarios. I consider modeling human behavior and predicting their decisions to be necessary for facing these challenges, as well as the development of formal models for the agent’s decision-making. Thus, my research studies build on methods and algorithms from Machine Learning, Decision Theory and Game Theory, non-classical logic, optimization under uncertainty and psychology.


1. Samuel Barrett, Peter Stone, Sarit Kraus, Avi Rosenfeld: Learning Teammate Models for Ad Hoc Teamwork. Proc. of ALA workshop (2012)
2. Michael Wooldridge, Sarit Kraus: Delegating Decisions in Strategic Settings Authors. Proc. of ECAI 2012
3. Maier Fenster, Inon Zuckerman, Sarit Kraus: Guiding User Choice During Discussion by Silence, Examples and Justifications. Proc. of ECAI (2012
4. James Pita, Richard John, Rajiv Maheswaran, Milind Tambe, Sarit Kraus: A Robust Approach to Addressing Human Adversaries in Security Games. Proc. of ECAI (2012)
5. Asaf Frieder, Raz Lin, Sarit Kraus: Agent-human Coordination with Communication Costs under Uncertainty. Proc of AAAI 2012


1. Computers Arguing with People (CAP)
2. Advanced Virtual Human (AVH)
3. The LAW-TRAIN project (LAW-TRAIN)