Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912690540> ?p ?o ?g. }
- W2912690540 abstract "Stochastic approximation (SA) is a key method used in statistical learning. Recently, its non-asymptotic convergence analysis has been considered in many papers. However, most of the prior analyses are made under restrictive assumptions such as unbiased gradient estimates and convex objective function, which significantly limit their applications to sophisticated tasks such as online and reinforcement learning. These restrictions are all essentially relaxed in this work. In particular, we analyze a general SA scheme to minimize a non-convex, smooth objective function. We consider update procedure whose drift term depends on a state-dependent Markov chain and the mean field is not necessarily of gradient type, covering approximate second-order method and allowing asymptotic bias for the one-step updates. We illustrate these settings with the online EM algorithm and the policy-gradient method for average reward maximization in reinforcement learning." @default.
- W2912690540 created "2019-02-21" @default.
- W2912690540 creator A5000295817 @default.
- W2912690540 creator A5015585821 @default.
- W2912690540 creator A5028611632 @default.
- W2912690540 creator A5055166851 @default.
- W2912690540 date "2019-02-02" @default.
- W2912690540 modified "2023-09-27" @default.
- W2912690540 title "Non-asymptotic Analysis of Biased Stochastic Approximation Scheme" @default.
- W2912690540 cites W1499021337 @default.
- W2912690540 cites W1568229137 @default.
- W2912690540 cites W1686266550 @default.
- W2912690540 cites W1990585910 @default.
- W2912690540 cites W1994616650 @default.
- W2912690540 cites W2009303086 @default.
- W2912690540 cites W2049633694 @default.
- W2912690540 cites W2053742104 @default.
- W2912690540 cites W2094364653 @default.
- W2912690540 cites W2119717200 @default.
- W2912690540 cites W2120717492 @default.
- W2912690540 cites W2121863487 @default.
- W2912690540 cites W2127684734 @default.
- W2912690540 cites W2132211083 @default.
- W2912690540 cites W2155027007 @default.
- W2912690540 cites W2156779765 @default.
- W2912690540 cites W2165131254 @default.
- W2912690540 cites W2185602187 @default.
- W2912690540 cites W2591423585 @default.
- W2912690540 cites W2796986421 @default.
- W2912690540 cites W2798888671 @default.
- W2912690540 cites W2886474253 @default.
- W2912690540 cites W2892339771 @default.
- W2912690540 cites W2953334758 @default.
- W2912690540 cites W2962737134 @default.
- W2912690540 cites W2963411541 @default.
- W2912690540 cites W2963433607 @default.
- W2912690540 cites W2963470657 @default.
- W2912690540 cites W2963695783 @default.
- W2912690540 cites W2964073168 @default.
- W2912690540 cites W2964123095 @default.
- W2912690540 cites W3136903997 @default.
- W2912690540 cites W68072195 @default.
- W2912690540 cites W2804329956 @default.
- W2912690540 hasPublicationYear "2019" @default.
- W2912690540 type Work @default.
- W2912690540 sameAs 2912690540 @default.
- W2912690540 citedByCount "11" @default.
- W2912690540 countsByYear W29126905402019 @default.
- W2912690540 countsByYear W29126905402020 @default.
- W2912690540 countsByYear W29126905402023 @default.
- W2912690540 crossrefType "posted-content" @default.
- W2912690540 hasAuthorship W2912690540A5000295817 @default.
- W2912690540 hasAuthorship W2912690540A5015585821 @default.
- W2912690540 hasAuthorship W2912690540A5028611632 @default.
- W2912690540 hasAuthorship W2912690540A5055166851 @default.
- W2912690540 hasConcept C105795698 @default.
- W2912690540 hasConcept C111110010 @default.
- W2912690540 hasConcept C112680207 @default.
- W2912690540 hasConcept C121332964 @default.
- W2912690540 hasConcept C126255220 @default.
- W2912690540 hasConcept C134306372 @default.
- W2912690540 hasConcept C14036430 @default.
- W2912690540 hasConcept C145446738 @default.
- W2912690540 hasConcept C151201525 @default.
- W2912690540 hasConcept C154945302 @default.
- W2912690540 hasConcept C157972887 @default.
- W2912690540 hasConcept C162324750 @default.
- W2912690540 hasConcept C205147927 @default.
- W2912690540 hasConcept C2524010 @default.
- W2912690540 hasConcept C26517878 @default.
- W2912690540 hasConcept C2776330181 @default.
- W2912690540 hasConcept C2777303404 @default.
- W2912690540 hasConcept C28826006 @default.
- W2912690540 hasConcept C33923547 @default.
- W2912690540 hasConcept C38652104 @default.
- W2912690540 hasConcept C41008148 @default.
- W2912690540 hasConcept C50522688 @default.
- W2912690540 hasConcept C55479107 @default.
- W2912690540 hasConcept C61797465 @default.
- W2912690540 hasConcept C62520636 @default.
- W2912690540 hasConcept C78458016 @default.
- W2912690540 hasConcept C86803240 @default.
- W2912690540 hasConcept C97541855 @default.
- W2912690540 hasConcept C98763669 @default.
- W2912690540 hasConceptScore W2912690540C105795698 @default.
- W2912690540 hasConceptScore W2912690540C111110010 @default.
- W2912690540 hasConceptScore W2912690540C112680207 @default.
- W2912690540 hasConceptScore W2912690540C121332964 @default.
- W2912690540 hasConceptScore W2912690540C126255220 @default.
- W2912690540 hasConceptScore W2912690540C134306372 @default.
- W2912690540 hasConceptScore W2912690540C14036430 @default.
- W2912690540 hasConceptScore W2912690540C145446738 @default.
- W2912690540 hasConceptScore W2912690540C151201525 @default.
- W2912690540 hasConceptScore W2912690540C154945302 @default.
- W2912690540 hasConceptScore W2912690540C157972887 @default.
- W2912690540 hasConceptScore W2912690540C162324750 @default.
- W2912690540 hasConceptScore W2912690540C205147927 @default.
- W2912690540 hasConceptScore W2912690540C2524010 @default.
- W2912690540 hasConceptScore W2912690540C26517878 @default.
- W2912690540 hasConceptScore W2912690540C2776330181 @default.