Matches in SemOpenAlex for { <https://semopenalex.org/work/W4281739486> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4281739486 abstract "We study the problem of finding the Nash equilibrium in a two-player zero-sum Markov game. Due to its formulation as a minimax optimization program, a natural approach to solve the problem is to perform gradient descent/ascent with respect to each player in an alternating fashion. However, due to the non-convexity/non-concavity of the underlying objective function, theoretical understandings of this method are limited. In our paper, we consider solving an entropy-regularized variant of the Markov game. The regularization introduces structure into the optimization landscape that make the solutions more identifiable and allow the problem to be solved more efficiently. Our main contribution is to show that under proper choices of the regularization parameter, the gradient descent ascent algorithm converges to the Nash equilibrium of the original unregularized problem. We explicitly characterize the finite-time performance of the last iterate of our algorithm, which vastly improves over the existing convergence bound of the gradient descent ascent algorithm without regularization. Finally, we complement the analysis with numerical simulations that illustrate the accelerated convergence of the algorithm." @default.
- W4281739486 created "2022-06-13" @default.
- W4281739486 creator A5035207859 @default.
- W4281739486 creator A5041443633 @default.
- W4281739486 creator A5069872754 @default.
- W4281739486 date "2022-05-26" @default.
- W4281739486 modified "2023-09-27" @default.
- W4281739486 title "Regularized Gradient Descent Ascent for Two-Player Zero-Sum Markov Games" @default.
- W4281739486 doi "https://doi.org/10.48550/arxiv.2205.13746" @default.
- W4281739486 hasPublicationYear "2022" @default.
- W4281739486 type Work @default.
- W4281739486 citedByCount "0" @default.
- W4281739486 crossrefType "posted-content" @default.
- W4281739486 hasAuthorship W4281739486A5035207859 @default.
- W4281739486 hasAuthorship W4281739486A5041443633 @default.
- W4281739486 hasAuthorship W4281739486A5069872754 @default.
- W4281739486 hasBestOaLocation W42817394861 @default.
- W4281739486 hasConcept C105795698 @default.
- W4281739486 hasConcept C106159729 @default.
- W4281739486 hasConcept C115680565 @default.
- W4281739486 hasConcept C126255220 @default.
- W4281739486 hasConcept C136356330 @default.
- W4281739486 hasConcept C137836250 @default.
- W4281739486 hasConcept C149728462 @default.
- W4281739486 hasConcept C153258448 @default.
- W4281739486 hasConcept C154945302 @default.
- W4281739486 hasConcept C162324750 @default.
- W4281739486 hasConcept C2776135515 @default.
- W4281739486 hasConcept C28826006 @default.
- W4281739486 hasConcept C33923547 @default.
- W4281739486 hasConcept C41008148 @default.
- W4281739486 hasConcept C46814582 @default.
- W4281739486 hasConcept C50644808 @default.
- W4281739486 hasConcept C72134830 @default.
- W4281739486 hasConcept C98763669 @default.
- W4281739486 hasConceptScore W4281739486C105795698 @default.
- W4281739486 hasConceptScore W4281739486C106159729 @default.
- W4281739486 hasConceptScore W4281739486C115680565 @default.
- W4281739486 hasConceptScore W4281739486C126255220 @default.
- W4281739486 hasConceptScore W4281739486C136356330 @default.
- W4281739486 hasConceptScore W4281739486C137836250 @default.
- W4281739486 hasConceptScore W4281739486C149728462 @default.
- W4281739486 hasConceptScore W4281739486C153258448 @default.
- W4281739486 hasConceptScore W4281739486C154945302 @default.
- W4281739486 hasConceptScore W4281739486C162324750 @default.
- W4281739486 hasConceptScore W4281739486C2776135515 @default.
- W4281739486 hasConceptScore W4281739486C28826006 @default.
- W4281739486 hasConceptScore W4281739486C33923547 @default.
- W4281739486 hasConceptScore W4281739486C41008148 @default.
- W4281739486 hasConceptScore W4281739486C46814582 @default.
- W4281739486 hasConceptScore W4281739486C50644808 @default.
- W4281739486 hasConceptScore W4281739486C72134830 @default.
- W4281739486 hasConceptScore W4281739486C98763669 @default.
- W4281739486 hasLocation W42817394861 @default.
- W4281739486 hasOpenAccess W4281739486 @default.
- W4281739486 hasPrimaryLocation W42817394861 @default.
- W4281739486 hasRelatedWork W1555584805 @default.
- W4281739486 hasRelatedWork W1981843668 @default.
- W4281739486 hasRelatedWork W2582724188 @default.
- W4281739486 hasRelatedWork W2963133087 @default.
- W4281739486 hasRelatedWork W3015030348 @default.
- W4281739486 hasRelatedWork W4221167780 @default.
- W4281739486 hasRelatedWork W4281739486 @default.
- W4281739486 hasRelatedWork W4285884497 @default.
- W4281739486 hasRelatedWork W4308234978 @default.
- W4281739486 hasRelatedWork W4312397237 @default.
- W4281739486 isParatext "false" @default.
- W4281739486 isRetracted "false" @default.
- W4281739486 workType "article" @default.