Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226025870> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4226025870 abstract "Stochastic approximation, a data-driven approach for finding the root of an unknown operator, provides a unified framework for solving many problems in stochastic optimization and reinforcement learning. Motivated by a growing interest in multi-agent and multi-task learning, we study a decentralized variant of stochastic approximation over a network of agents, where the goal is to find the root of the aggregate of the local operators at the agents. In this method, each agent implements a local stochastic approximation using noisy samples from its operator while averaging its iterates with the ones received from its neighbors. Our main contribution is to provide a finite-time analysis of the decentralized stochastic approximation method and to characterize the impacts of the underlying communication topology between agents. Our model for the data observed at each agent is that it is sampled from a Markov process; this lack of independence makes the iterates biased and (potentially) unbounded. Under mild assumptions we show that the convergence rate of the proposed method is essentially the same as if the samples were independent, differing only by a log factor that represents the mixing time of the Markov process. Finally, we present applications of the proposed method on a number of interesting learning problems in multi-agent systems, including distributed robust system identification and decentralized Q-learning for solving multitask reinforcement learning." @default.
- W4226025870 created "2022-05-05" @default.
- W4226025870 creator A5035207859 @default.
- W4226025870 creator A5041443633 @default.
- W4226025870 creator A5069872754 @default.
- W4226025870 date "2021-12-14" @default.
- W4226025870 modified "2023-09-24" @default.
- W4226025870 title "Finite-Time Analysis of Decentralized Stochastic Approximation with Applications in Multi-Agent and Multi-Task Learning" @default.
- W4226025870 cites W1616857247 @default.
- W4226025870 cites W1979043738 @default.
- W4226025870 cites W2081756148 @default.
- W4226025870 cites W2150123286 @default.
- W4226025870 cites W2885549115 @default.
- W4226025870 cites W2912388896 @default.
- W4226025870 cites W2962771678 @default.
- W4226025870 cites W2964026281 @default.
- W4226025870 cites W3041202696 @default.
- W4226025870 cites W3118491379 @default.
- W4226025870 cites W3132503596 @default.
- W4226025870 cites W32403112 @default.
- W4226025870 doi "https://doi.org/10.1109/cdc45484.2021.9683363" @default.
- W4226025870 hasPublicationYear "2021" @default.
- W4226025870 type Work @default.
- W4226025870 citedByCount "1" @default.
- W4226025870 countsByYear W42260258702023 @default.
- W4226025870 crossrefType "proceedings-article" @default.
- W4226025870 hasAuthorship W4226025870A5035207859 @default.
- W4226025870 hasAuthorship W4226025870A5041443633 @default.
- W4226025870 hasAuthorship W4226025870A5069872754 @default.
- W4226025870 hasConcept C104317684 @default.
- W4226025870 hasConcept C105795698 @default.
- W4226025870 hasConcept C119857082 @default.
- W4226025870 hasConcept C126255220 @default.
- W4226025870 hasConcept C134306372 @default.
- W4226025870 hasConcept C140479938 @default.
- W4226025870 hasConcept C154945302 @default.
- W4226025870 hasConcept C158448853 @default.
- W4226025870 hasConcept C159886148 @default.
- W4226025870 hasConcept C17020691 @default.
- W4226025870 hasConcept C185592680 @default.
- W4226025870 hasConcept C26517878 @default.
- W4226025870 hasConcept C33923547 @default.
- W4226025870 hasConcept C35651441 @default.
- W4226025870 hasConcept C38652104 @default.
- W4226025870 hasConcept C41008148 @default.
- W4226025870 hasConcept C41550386 @default.
- W4226025870 hasConcept C55479107 @default.
- W4226025870 hasConcept C55493867 @default.
- W4226025870 hasConcept C8272713 @default.
- W4226025870 hasConcept C86339819 @default.
- W4226025870 hasConcept C97541855 @default.
- W4226025870 hasConcept C98763669 @default.
- W4226025870 hasConceptScore W4226025870C104317684 @default.
- W4226025870 hasConceptScore W4226025870C105795698 @default.
- W4226025870 hasConceptScore W4226025870C119857082 @default.
- W4226025870 hasConceptScore W4226025870C126255220 @default.
- W4226025870 hasConceptScore W4226025870C134306372 @default.
- W4226025870 hasConceptScore W4226025870C140479938 @default.
- W4226025870 hasConceptScore W4226025870C154945302 @default.
- W4226025870 hasConceptScore W4226025870C158448853 @default.
- W4226025870 hasConceptScore W4226025870C159886148 @default.
- W4226025870 hasConceptScore W4226025870C17020691 @default.
- W4226025870 hasConceptScore W4226025870C185592680 @default.
- W4226025870 hasConceptScore W4226025870C26517878 @default.
- W4226025870 hasConceptScore W4226025870C33923547 @default.
- W4226025870 hasConceptScore W4226025870C35651441 @default.
- W4226025870 hasConceptScore W4226025870C38652104 @default.
- W4226025870 hasConceptScore W4226025870C41008148 @default.
- W4226025870 hasConceptScore W4226025870C41550386 @default.
- W4226025870 hasConceptScore W4226025870C55479107 @default.
- W4226025870 hasConceptScore W4226025870C55493867 @default.
- W4226025870 hasConceptScore W4226025870C8272713 @default.
- W4226025870 hasConceptScore W4226025870C86339819 @default.
- W4226025870 hasConceptScore W4226025870C97541855 @default.
- W4226025870 hasConceptScore W4226025870C98763669 @default.
- W4226025870 hasLocation W42260258701 @default.
- W4226025870 hasOpenAccess W4226025870 @default.
- W4226025870 hasPrimaryLocation W42260258701 @default.
- W4226025870 hasRelatedWork W1992917539 @default.
- W4226025870 hasRelatedWork W2071291306 @default.
- W4226025870 hasRelatedWork W2352247930 @default.
- W4226025870 hasRelatedWork W2602600938 @default.
- W4226025870 hasRelatedWork W2971059638 @default.
- W4226025870 hasRelatedWork W3123631523 @default.
- W4226025870 hasRelatedWork W3143470420 @default.
- W4226025870 hasRelatedWork W4226025870 @default.
- W4226025870 hasRelatedWork W4287241112 @default.
- W4226025870 hasRelatedWork W1514028002 @default.
- W4226025870 isParatext "false" @default.
- W4226025870 isRetracted "false" @default.
- W4226025870 workType "article" @default.