b. 1953
Joseph Halpern is a computer scientist and logician at Cornell University whose work bridges formal epistemology, artificial intelligence, and game theory. He is best known for foundational contributions to the logic of knowledge and belief, probabilistic reasoning, and the formal analysis of causality. His research has shaped how both philosophers and computer scientists model what agents know, how they update beliefs, and how causal responsibility is assigned.
Co-developed the possible-worlds framework for reasoning about distributed knowledge (with Fagin, Moses, and Vardi), foundational to epistemic logic in CS
Developed the Halpern-Pearl structural causal model of actual causation, widely adopted in philosophy and AI
Co-authored 'Reasoning About Knowledge' (1995), the standard reference on epistemic logic for computer scientists
Contributed formal analyses of plausibility measures and conditional probability for belief revision and game theory
Demonstrated limits on agents' logical omniscience, challenging the a priori status of logical knowledge in resource-bounded settings
Plausibility updates in sequential games during actual play differ in interpretation from plausibility updates used in pregame deliberation for Backward Induction.
claimThere is a fundamental tension between treating logical knowledge as a priori and the computational intractability of deciding logical validity.
Plausibility updates in sequential games during actual play differ in interpretation from plausibility updates used in pregame deliberation for Backward Induction.
claimThere is a fundamental tension between treating logical knowledge as a priori and the computational intractability of deciding logical validity.