Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197697946> ?p ?o ?g. }
- W3197697946 abstract "Adaptive gradient methods including Adam, AdaGrad, and their variants have been very successful for training deep learning models, such as neural networks. Meanwhile, given the need for distributed computing, distributed optimization algorithms are rapidly becoming a focal point. With the growth of computing power and the need for using machine learning models on mobile devices, the communication cost of distributed training algorithms needs careful consideration. In this paper, we introduce novel convergent decentralized adaptive gradient methods and rigorously incorporate adaptive gradient methods into decentralized training procedures. Specifically, we propose a general algorithmic framework that can convert existing adaptive gradient methods to their decentralized counterparts. In addition, we thoroughly analyze the convergence behavior of the proposed algorithmic framework and show that if a given adaptive gradient method converges, under some specific conditions, then its decentralized counterpart is also convergent. We illustrate the benefit of our generic decentralized framework on a prototype method, i.e., AMSGrad, both theoretically and numerically." @default.
- W3197697946 created "2021-09-13" @default.
- W3197697946 creator A5015585821 @default.
- W3197697946 creator A5049896980 @default.
- W3197697946 creator A5073070877 @default.
- W3197697946 creator A5075541712 @default.
- W3197697946 date "2021-09-07" @default.
- W3197697946 modified "2023-10-16" @default.
- W3197697946 title "On the Convergence of Decentralized Adaptive Gradient Methods." @default.
- W3197697946 cites W1442374986 @default.
- W3197697946 cites W1571416372 @default.
- W3197697946 cites W1616857247 @default.
- W3197697946 cites W1994616650 @default.
- W3197697946 cites W2044212084 @default.
- W3197697946 cites W2111006384 @default.
- W3197697946 cites W2119913432 @default.
- W3197697946 cites W2124509324 @default.
- W3197697946 cites W2134711723 @default.
- W3197697946 cites W2146502635 @default.
- W3197697946 cites W2164278908 @default.
- W3197697946 cites W2266148951 @default.
- W3197697946 cites W2541884796 @default.
- W3197697946 cites W2733300693 @default.
- W3197697946 cites W2769644379 @default.
- W3197697946 cites W2785523195 @default.
- W3197697946 cites W2789243870 @default.
- W3197697946 cites W2808402975 @default.
- W3197697946 cites W2886294525 @default.
- W3197697946 cites W2886463271 @default.
- W3197697946 cites W2890924858 @default.
- W3197697946 cites W2891952073 @default.
- W3197697946 cites W2902410171 @default.
- W3197697946 cites W2907225497 @default.
- W3197697946 cites W2908334250 @default.
- W3197697946 cites W2911327153 @default.
- W3197697946 cites W2945586457 @default.
- W3197697946 cites W2946511237 @default.
- W3197697946 cites W2946815927 @default.
- W3197697946 cites W2954270931 @default.
- W3197697946 cites W2963228337 @default.
- W3197697946 cites W2963563140 @default.
- W3197697946 cites W2963613486 @default.
- W3197697946 cites W2963803379 @default.
- W3197697946 cites W2964121744 @default.
- W3197697946 cites W2964267428 @default.
- W3197697946 cites W3037774634 @default.
- W3197697946 cites W3094279629 @default.
- W3197697946 cites W3101036738 @default.
- W3197697946 cites W3124442241 @default.
- W3197697946 cites W3173839890 @default.
- W3197697946 hasPublicationYear "2021" @default.
- W3197697946 type Work @default.
- W3197697946 sameAs 3197697946 @default.
- W3197697946 citedByCount "2" @default.
- W3197697946 countsByYear W31976979462021 @default.
- W3197697946 crossrefType "posted-content" @default.
- W3197697946 hasAuthorship W3197697946A5015585821 @default.
- W3197697946 hasAuthorship W3197697946A5049896980 @default.
- W3197697946 hasAuthorship W3197697946A5073070877 @default.
- W3197697946 hasAuthorship W3197697946A5075541712 @default.
- W3197697946 hasConcept C11413529 @default.
- W3197697946 hasConcept C115680565 @default.
- W3197697946 hasConcept C126255220 @default.
- W3197697946 hasConcept C154945302 @default.
- W3197697946 hasConcept C162324750 @default.
- W3197697946 hasConcept C2524010 @default.
- W3197697946 hasConcept C2777303404 @default.
- W3197697946 hasConcept C28719098 @default.
- W3197697946 hasConcept C33923547 @default.
- W3197697946 hasConcept C41008148 @default.
- W3197697946 hasConcept C50522688 @default.
- W3197697946 hasConcept C50644808 @default.
- W3197697946 hasConceptScore W3197697946C11413529 @default.
- W3197697946 hasConceptScore W3197697946C115680565 @default.
- W3197697946 hasConceptScore W3197697946C126255220 @default.
- W3197697946 hasConceptScore W3197697946C154945302 @default.
- W3197697946 hasConceptScore W3197697946C162324750 @default.
- W3197697946 hasConceptScore W3197697946C2524010 @default.
- W3197697946 hasConceptScore W3197697946C2777303404 @default.
- W3197697946 hasConceptScore W3197697946C28719098 @default.
- W3197697946 hasConceptScore W3197697946C33923547 @default.
- W3197697946 hasConceptScore W3197697946C41008148 @default.
- W3197697946 hasConceptScore W3197697946C50522688 @default.
- W3197697946 hasConceptScore W3197697946C50644808 @default.
- W3197697946 hasLocation W31976979461 @default.
- W3197697946 hasOpenAccess W3197697946 @default.
- W3197697946 hasPrimaryLocation W31976979461 @default.
- W3197697946 hasRelatedWork W1538051798 @default.
- W3197697946 hasRelatedWork W1543977710 @default.
- W3197697946 hasRelatedWork W1968938284 @default.
- W3197697946 hasRelatedWork W2074129780 @default.
- W3197697946 hasRelatedWork W2101106169 @default.
- W3197697946 hasRelatedWork W2317856971 @default.
- W3197697946 hasRelatedWork W2885584824 @default.
- W3197697946 hasRelatedWork W2914840043 @default.
- W3197697946 hasRelatedWork W2958888782 @default.
- W3197697946 hasRelatedWork W2966868123 @default.
- W3197697946 hasRelatedWork W2977225700 @default.
- W3197697946 hasRelatedWork W3007468665 @default.
- W3197697946 hasRelatedWork W3033405336 @default.