Limit value of dynamic zero-sum games with vanishing stage duration
Sylvain Sorin (IMJ)

TL;DR
This paper investigates the limiting behavior of dynamic zero-sum games with vanishing stage durations, establishing the existence of the limit value as the partition mesh approaches zero across various game types.
Contribution
It proves the existence of the limit value for dynamic zero-sum games with continuous-time Markov controls as the stage duration shrinks, extending to stochastic and symmetric no-information cases.
Findings
Limit value exists as partition mesh approaches zero.
Results apply to stochastic and symmetric no-information games.
Analysis reduces to a deterministic differential game.
Abstract
We consider two person zero-sum games where the players control, at discrete times {tn} induced by a partition of R + , a continuous time Markov state process. We prove that the limit of the values v exist as the mesh of goes to 0. The analysis covers the cases of : 1) stochastic games (where both players know the state) 2) symmetric no information. The proof is by reduction to a deterministic differential game.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGame Theory and Applications · Game Theory and Voting Systems · Reinforcement Learning in Robotics
