Rational decisions in non-probabilistic settings

  • Sergei Artemov | Graduate Center of the City University of New York

The knowledge-based rational decision model (KBR-model) is built on explicit epistemic reading of standard game-theoretical assumptions, e.g., Harsanyi’s Maximin Postulate, in a non-probabilistic setting. This model suggests following maximin strategy over all scenarios which the agent considers possible to the best of his knowledge.

In this talk, we show that KBR is the only non-probabilistic decision making method which is definitive, rational, and based exclusively on knowledge. We analyze other decision methods: Nash equilibrium and subgame perfect equilibrium, backward induction solution, pure maximin, eliminating dominated strategies, and for each of these cases indicate which of three properties (definitive, rational, or based exclusively on knowledge) fails.

Speaker Details

Sergei Artemov is a Distinguished Professor of Computer Science, the founder and head of the Research Laboratory for Logic and Computation at the Graduate Center of the City University of New York. Professor Artemov’s professional interests are logic in computer science, epistemology, game theory, automated deduction and verification, and optimal control and hybrid systems. He has developed a Justification Logic that incorporates justifications into epistemic logic and renders its new, evidence-based foundation. His most recent project is in Game Theory where he has developed a theory of knowledge-based rational decisions in non-probabilistic settings.Professor Artemov has authored 140 research papers and supervised 21 Ph.D. dissertations. He is an editor of several journals and the organizer of a number of international conferences, including the symposium series Logical Foundations of Computer Science. He has delivered a Distinguished Lecture for the New York Academy of Sciences, Clifford Lectures, the Spinoza Lecture for the European Association for Computer Science Logic, the keynote lecture for the Kurt Gödel Society in Vienna.

    • Portrait of Jeff Running

      Jeff Running