SemOpenAlex |

SemOpenAlex

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3081034653> ?p ?o ?g. }

Showing items 1 to 87 of 87 with 100 items per page.

W3081034653 abstract "Client devices at the edge are generating increasingly large amounts of rich data suitable for learning powerful statistical models. However, privacy concerns and heavy communication load make it infeasible to move the client data to a centralized location for training. In many distributed learning setups, client nodes carry out gradient computations on their local data while the central master server receives the local gradients and aggregates them to take the global model update step. To guarantee robustness against straggling communication links, we consider a hierarchical setup with n e clients and n h reliable helper nodes that are available to aid in gradient aggregation at the master. To achieve resiliency against straggling client-to-helpers links, we propose two approaches leveraging coded redundancy. First is the Aligned Repetition Coding (ARC) that repeats gradient components on the helper links, allowing significant partial aggregations at the helpers, resulting in a helpers-to-master communication load (C HM ) of O(n h ). ARC however results in a client-to-helpers communication load (C EH ) of Θ(n h ), which is prohibitive for client nodes due to limited and costly bandwidth. We thus propose Aligned Minimum Distance Separable Coding (AMC) that achieves optimal C EH of Θ(1) for a given resiliency threshold by using MDS code over the gradient components, while achieving a C HM of O(n e )." @default.
W3081034653 created "2020-09-01" @default.
W3081034653 creator A5019288145 @default.
W3081034653 creator A5040270189 @default.
W3081034653 creator A5047191296 @default.
W3081034653 creator A5059780399 @default.
W3081034653 date "2020-06-01" @default.
W3081034653 modified "2023-09-26" @default.
W3081034653 title "Hierarchical Coded Gradient Aggregation for Learning at the Edge" @default.
W3081034653 cites W1587463421 @default.
W3081034653 cites W2108603792 @default.
W3081034653 cites W2568772110 @default.
W3081034653 cites W2790718346 @default.
W3081034653 cites W2886711970 @default.
W3081034653 cites W2916505310 @default.
W3081034653 cites W2922342388 @default.
W3081034653 cites W2943865028 @default.
W3081034653 cites W2962884006 @default.
W3081034653 cites W2962904485 @default.
W3081034653 cites W2963062115 @default.
W3081034653 cites W2963726510 @default.
W3081034653 cites W2963837493 @default.
W3081034653 cites W2964120628 @default.
W3081034653 cites W2971544778 @default.
W3081034653 cites W2976436958 @default.
W3081034653 cites W2976668007 @default.
W3081034653 cites W2976864125 @default.
W3081034653 cites W2977140747 @default.
W3081034653 cites W2985805285 @default.
W3081034653 cites W2987490938 @default.
W3081034653 cites W3009132791 @default.
W3081034653 cites W3011729304 @default.
W3081034653 cites W3081034653 @default.
W3081034653 doi "https://doi.org/10.1109/isit44484.2020.9174077" @default.
W3081034653 hasPublicationYear "2020" @default.
W3081034653 type Work @default.
W3081034653 sameAs 3081034653 @default.
W3081034653 citedByCount "12" @default.
W3081034653 countsByYear W30810346532019 @default.
W3081034653 countsByYear W30810346532020 @default.
W3081034653 countsByYear W30810346532021 @default.
W3081034653 countsByYear W30810346532022 @default.
W3081034653 countsByYear W30810346532023 @default.
W3081034653 crossrefType "proceedings-article" @default.
W3081034653 hasAuthorship W3081034653A5019288145 @default.
W3081034653 hasAuthorship W3081034653A5040270189 @default.
W3081034653 hasAuthorship W3081034653A5047191296 @default.
W3081034653 hasAuthorship W3081034653A5059780399 @default.
W3081034653 hasConcept C104317684 @default.
W3081034653 hasConcept C111919701 @default.
W3081034653 hasConcept C152124472 @default.
W3081034653 hasConcept C154945302 @default.
W3081034653 hasConcept C41008148 @default.
W3081034653 hasConcept C55493867 @default.
W3081034653 hasConcept C63479239 @default.
W3081034653 hasConcept C7545210 @default.
W3081034653 hasConcept C77088390 @default.
W3081034653 hasConcept C80444323 @default.
W3081034653 hasConcept C86803240 @default.
W3081034653 hasConceptScore W3081034653C104317684 @default.
W3081034653 hasConceptScore W3081034653C111919701 @default.
W3081034653 hasConceptScore W3081034653C152124472 @default.
W3081034653 hasConceptScore W3081034653C154945302 @default.
W3081034653 hasConceptScore W3081034653C41008148 @default.
W3081034653 hasConceptScore W3081034653C55493867 @default.
W3081034653 hasConceptScore W3081034653C63479239 @default.
W3081034653 hasConceptScore W3081034653C7545210 @default.
W3081034653 hasConceptScore W3081034653C77088390 @default.
W3081034653 hasConceptScore W3081034653C80444323 @default.
W3081034653 hasConceptScore W3081034653C86803240 @default.
W3081034653 hasLocation W30810346531 @default.
W3081034653 hasOpenAccess W3081034653 @default.
W3081034653 hasPrimaryLocation W30810346531 @default.
W3081034653 hasRelatedWork W1543599621 @default.
W3081034653 hasRelatedWork W2002428578 @default.
W3081034653 hasRelatedWork W2035976912 @default.
W3081034653 hasRelatedWork W2541791370 @default.
W3081034653 hasRelatedWork W2775182575 @default.
W3081034653 hasRelatedWork W3094187672 @default.
W3081034653 hasRelatedWork W3164116745 @default.
W3081034653 hasRelatedWork W3213778687 @default.
W3081034653 hasRelatedWork W4379876687 @default.
W3081034653 hasRelatedWork W4385415357 @default.
W3081034653 isParatext "false" @default.
W3081034653 isRetracted "false" @default.
W3081034653 magId "3081034653" @default.
W3081034653 workType "article" @default.