Matches in SemOpenAlex for { <https://semopenalex.org/work/W3006493147> ?p ?o ?g. }
- W3006493147 abstract "We introduce a novel and efficient algorithm called the stochastic approximate gradient descent (SAGD), as an alternative to the stochastic gradient descent for cases where unbiased stochastic gradients cannot be trivially obtained. Traditional methods for such problems rely on general-purpose sampling techniques such as Markov chain Monte Carlo, which typically requires manual intervention for tuning parameters and does not work efficiently in practice. Instead, SAGD makes use of the Langevin algorithm to construct stochastic gradients that are biased in finite steps but accurate asymptotically, enabling us to theoretically establish the convergence guarantee for SAGD. Inspired by our theoretical analysis, we also provide useful guidelines for its practical implementation. Finally, we show that SAGD performs well experimentally in popular statistical and machine learning problems such as the expectation-maximization algorithm and the variational autoencoders." @default.
- W3006493147 created "2020-02-24" @default.
- W3006493147 creator A5039257426 @default.
- W3006493147 creator A5058010200 @default.
- W3006493147 date "2020-02-13" @default.
- W3006493147 modified "2023-10-16" @default.
- W3006493147 title "Stochastic Approximate Gradient Descent via the Langevin Algorithm" @default.
- W3006493147 cites W114517082 @default.
- W3006493147 cites W1499798934 @default.
- W3006493147 cites W1556278552 @default.
- W3006493147 cites W1983452151 @default.
- W3006493147 cites W1994616650 @default.
- W3006493147 cites W2020999234 @default.
- W3006493147 cites W2047834101 @default.
- W3006493147 cites W2049633694 @default.
- W3006493147 cites W2056760934 @default.
- W3006493147 cites W2094034395 @default.
- W3006493147 cites W2104049095 @default.
- W3006493147 cites W2107438106 @default.
- W3006493147 cites W2116825644 @default.
- W3006493147 cites W2130416410 @default.
- W3006493147 cites W2138309709 @default.
- W3006493147 cites W2146502635 @default.
- W3006493147 cites W2167433878 @default.
- W3006493147 cites W2186210550 @default.
- W3006493147 cites W2222154095 @default.
- W3006493147 cites W2323824333 @default.
- W3006493147 cites W2528137865 @default.
- W3006493147 cites W2788289427 @default.
- W3006493147 cites W2802698595 @default.
- W3006493147 cites W2883697110 @default.
- W3006493147 cites W2907225497 @default.
- W3006493147 cites W2907338582 @default.
- W3006493147 cites W2911398476 @default.
- W3006493147 cites W2962794482 @default.
- W3006493147 cites W2963275229 @default.
- W3006493147 cites W2963433607 @default.
- W3006493147 cites W2963965485 @default.
- W3006493147 cites W2964121744 @default.
- W3006493147 cites W2964301034 @default.
- W3006493147 cites W2972965199 @default.
- W3006493147 cites W2979473621 @default.
- W3006493147 cites W3109346590 @default.
- W3006493147 cites W3124025666 @default.
- W3006493147 cites W621546036 @default.
- W3006493147 cites W6908809 @default.
- W3006493147 cites W2072634211 @default.
- W3006493147 doi "https://doi.org/10.48550/arxiv.2002.05519" @default.
- W3006493147 hasPublicationYear "2020" @default.
- W3006493147 type Work @default.
- W3006493147 sameAs 3006493147 @default.
- W3006493147 citedByCount "0" @default.
- W3006493147 crossrefType "posted-content" @default.
- W3006493147 hasAuthorship W3006493147A5039257426 @default.
- W3006493147 hasAuthorship W3006493147A5058010200 @default.
- W3006493147 hasBestOaLocation W30064931471 @default.
- W3006493147 hasConcept C105795698 @default.
- W3006493147 hasConcept C107673813 @default.
- W3006493147 hasConcept C111350023 @default.
- W3006493147 hasConcept C11413529 @default.
- W3006493147 hasConcept C119857082 @default.
- W3006493147 hasConcept C126255220 @default.
- W3006493147 hasConcept C127413603 @default.
- W3006493147 hasConcept C146978453 @default.
- W3006493147 hasConcept C153258448 @default.
- W3006493147 hasConcept C154945302 @default.
- W3006493147 hasConcept C162324750 @default.
- W3006493147 hasConcept C19499675 @default.
- W3006493147 hasConcept C206688291 @default.
- W3006493147 hasConcept C2776637919 @default.
- W3006493147 hasConcept C2777303404 @default.
- W3006493147 hasConcept C2780004032 @default.
- W3006493147 hasConcept C28826006 @default.
- W3006493147 hasConcept C33923547 @default.
- W3006493147 hasConcept C41008148 @default.
- W3006493147 hasConcept C50522688 @default.
- W3006493147 hasConcept C50644808 @default.
- W3006493147 hasConcept C98763669 @default.
- W3006493147 hasConceptScore W3006493147C105795698 @default.
- W3006493147 hasConceptScore W3006493147C107673813 @default.
- W3006493147 hasConceptScore W3006493147C111350023 @default.
- W3006493147 hasConceptScore W3006493147C11413529 @default.
- W3006493147 hasConceptScore W3006493147C119857082 @default.
- W3006493147 hasConceptScore W3006493147C126255220 @default.
- W3006493147 hasConceptScore W3006493147C127413603 @default.
- W3006493147 hasConceptScore W3006493147C146978453 @default.
- W3006493147 hasConceptScore W3006493147C153258448 @default.
- W3006493147 hasConceptScore W3006493147C154945302 @default.
- W3006493147 hasConceptScore W3006493147C162324750 @default.
- W3006493147 hasConceptScore W3006493147C19499675 @default.
- W3006493147 hasConceptScore W3006493147C206688291 @default.
- W3006493147 hasConceptScore W3006493147C2776637919 @default.
- W3006493147 hasConceptScore W3006493147C2777303404 @default.
- W3006493147 hasConceptScore W3006493147C2780004032 @default.
- W3006493147 hasConceptScore W3006493147C28826006 @default.
- W3006493147 hasConceptScore W3006493147C33923547 @default.
- W3006493147 hasConceptScore W3006493147C41008148 @default.
- W3006493147 hasConceptScore W3006493147C50522688 @default.
- W3006493147 hasConceptScore W3006493147C50644808 @default.
- W3006493147 hasConceptScore W3006493147C98763669 @default.