Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048804513> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W3048804513 abstract "Although the distributed machine learning methods show the potential for the speed-up of training large deep neural networks, the communication cost has been the notorious bottleneck to constrain the performance. To address this challenge, the compression based communication-efficient distributed learning methods were designed to reduce the communication cost, and more recently the local error feedback was incorporated to compensate for the performance loss. However, in this paper, we will show the gradient mismatch problem of the local error feedback in centralized distributed training and this issue can lead to degraded performance compared with full-precision training. To solve this critical problem, we propose two novel techniques: 1) step ahead; 2) error averaging. Both our theoretical and empirical results show that our new methods can alleviate the gradient mismatch problem. Experiments show that we can even train textbf{faster with compressed gradient} than full-precision training textbf{regarding training epochs}." @default.
- W3048804513 created "2020-08-18" @default.
- W3048804513 creator A5060016795 @default.
- W3048804513 creator A5063099787 @default.
- W3048804513 creator A5069872293 @default.
- W3048804513 date "2020-08-13" @default.
- W3048804513 modified "2023-09-23" @default.
- W3048804513 title "Training Faster with Compressed Gradient" @default.
- W3048804513 cites W2117539524 @default.
- W3048804513 cites W2132737349 @default.
- W3048804513 cites W2194775991 @default.
- W3048804513 cites W2405578611 @default.
- W3048804513 cites W2407022425 @default.
- W3048804513 cites W2535838896 @default.
- W3048804513 cites W2769644379 @default.
- W3048804513 cites W2786602455 @default.
- W3048804513 cites W2789243870 @default.
- W3048804513 cites W2803965032 @default.
- W3048804513 cites W2889676205 @default.
- W3048804513 cites W2890924858 @default.
- W3048804513 cites W2911327153 @default.
- W3048804513 cites W2911863041 @default.
- W3048804513 cites W2962863496 @default.
- W3048804513 cites W2963179579 @default.
- W3048804513 cites W2963228337 @default.
- W3048804513 cites W2963766684 @default.
- W3048804513 cites W2963773265 @default.
- W3048804513 cites W2963803379 @default.
- W3048804513 cites W2963903325 @default.
- W3048804513 cites W2964004663 @default.
- W3048804513 cites W2965234876 @default.
- W3048804513 cites W2970432126 @default.
- W3048804513 cites W2970971581 @default.
- W3048804513 cites W2971064744 @default.
- W3048804513 cites W3037875189 @default.
- W3048804513 cites W3101036738 @default.
- W3048804513 cites W3118608800 @default.
- W3048804513 hasPublicationYear "2020" @default.
- W3048804513 type Work @default.
- W3048804513 sameAs 3048804513 @default.
- W3048804513 citedByCount "1" @default.
- W3048804513 countsByYear W30488045132021 @default.
- W3048804513 crossrefType "posted-content" @default.
- W3048804513 hasAuthorship W3048804513A5060016795 @default.
- W3048804513 hasAuthorship W3048804513A5063099787 @default.
- W3048804513 hasAuthorship W3048804513A5069872293 @default.
- W3048804513 hasConcept C113775141 @default.
- W3048804513 hasConcept C11413529 @default.
- W3048804513 hasConcept C121332964 @default.
- W3048804513 hasConcept C149635348 @default.
- W3048804513 hasConcept C153294291 @default.
- W3048804513 hasConcept C154945302 @default.
- W3048804513 hasConcept C2777211547 @default.
- W3048804513 hasConcept C2780513914 @default.
- W3048804513 hasConcept C2984842247 @default.
- W3048804513 hasConcept C41008148 @default.
- W3048804513 hasConcept C50644808 @default.
- W3048804513 hasConceptScore W3048804513C113775141 @default.
- W3048804513 hasConceptScore W3048804513C11413529 @default.
- W3048804513 hasConceptScore W3048804513C121332964 @default.
- W3048804513 hasConceptScore W3048804513C149635348 @default.
- W3048804513 hasConceptScore W3048804513C153294291 @default.
- W3048804513 hasConceptScore W3048804513C154945302 @default.
- W3048804513 hasConceptScore W3048804513C2777211547 @default.
- W3048804513 hasConceptScore W3048804513C2780513914 @default.
- W3048804513 hasConceptScore W3048804513C2984842247 @default.
- W3048804513 hasConceptScore W3048804513C41008148 @default.
- W3048804513 hasConceptScore W3048804513C50644808 @default.
- W3048804513 hasLocation W30488045131 @default.
- W3048804513 hasOpenAccess W3048804513 @default.
- W3048804513 hasPrimaryLocation W30488045131 @default.
- W3048804513 hasRelatedWork W2617019596 @default.
- W3048804513 hasRelatedWork W2617766261 @default.
- W3048804513 hasRelatedWork W2896404430 @default.
- W3048804513 hasRelatedWork W2901028208 @default.
- W3048804513 hasRelatedWork W2943847545 @default.
- W3048804513 hasRelatedWork W2952369090 @default.
- W3048804513 hasRelatedWork W2962710991 @default.
- W3048804513 hasRelatedWork W3087165870 @default.
- W3048804513 hasRelatedWork W3096004388 @default.
- W3048804513 hasRelatedWork W3100246930 @default.
- W3048804513 hasRelatedWork W3105148963 @default.
- W3048804513 hasRelatedWork W3118182806 @default.
- W3048804513 hasRelatedWork W3118543361 @default.
- W3048804513 hasRelatedWork W3131685502 @default.
- W3048804513 hasRelatedWork W3174871729 @default.
- W3048804513 hasRelatedWork W3198707348 @default.
- W3048804513 hasRelatedWork W3204450046 @default.
- W3048804513 hasRelatedWork W3204507145 @default.
- W3048804513 hasRelatedWork W3205035978 @default.
- W3048804513 hasRelatedWork W3212271975 @default.
- W3048804513 isParatext "false" @default.
- W3048804513 isRetracted "false" @default.
- W3048804513 magId "3048804513" @default.
- W3048804513 workType "article" @default.