Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312551440> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W4312551440 abstract "This paper considers the problem of communication and computation-efficient distributed learning via a wireless fading Multiple Access Channel (MAC). The distributed learning task is performed over a large network of nodes containing local data with the help of an edge server coordinating between the nodes. The information from each distributed node is transmitted as an analog signal through a noisy fading wireless MAC, using a common shaping waveform. The edge server receives a superposition of the analog signals, computes a new parameter estimate and communicates it back to the nodes, a process which continues until an appropriate convergence criterion is met. Unlike typical Federated learning approaches based on communication of local gradients and averaging at the edge server, in this paper, we investigate a scenario where the local nodes implement a second order optimization technique known as Determinantal Averaging. The communication complexity at each iteration per node of this method is the same as any gradient based method, i.e. <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$O(d)$</tex> , where <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$d$</tex> is the number of parameters. To reduce the computational load at each node, we also employ an approximate Newton method to compute the local Hessians. Under the usual assumptions of convexity and double differentiability on the local objective functions, we propose an algorithm titled Distributed Approximate Newton with Determinantal Averaging (DANDA). The state-of-art first and second-order distributed optimization algorithms are numerically compared with DANDA on a standard dataset with least squares based local objective functions (linear regression). Simulation results illustrate that DANDA not only displays faster convergence compared to gradient-based methods, but also compares favourably with exact distributed Newton methods, such as LocalNewton." @default.
- W4312551440 created "2023-01-05" @default.
- W4312551440 creator A5057431380 @default.
- W4312551440 creator A5091142549 @default.
- W4312551440 date "2022-09-12" @default.
- W4312551440 modified "2023-10-16" @default.
- W4312551440 title "On Analog Distributed Approximate Newton with Determinantal Averaging" @default.
- W4312551440 cites W2960514293 @default.
- W4312551440 cites W2969343398 @default.
- W4312551440 cites W2971330699 @default.
- W4312551440 cites W2981138228 @default.
- W4312551440 cites W3006919779 @default.
- W4312551440 cites W3036822879 @default.
- W4312551440 cites W3088234149 @default.
- W4312551440 cites W4225845194 @default.
- W4312551440 cites W4285142407 @default.
- W4312551440 doi "https://doi.org/10.1109/pimrc54779.2022.9977466" @default.
- W4312551440 hasPublicationYear "2022" @default.
- W4312551440 type Work @default.
- W4312551440 citedByCount "0" @default.
- W4312551440 crossrefType "proceedings-article" @default.
- W4312551440 hasAuthorship W4312551440A5057431380 @default.
- W4312551440 hasAuthorship W4312551440A5091142549 @default.
- W4312551440 hasConcept C106159729 @default.
- W4312551440 hasConcept C11413529 @default.
- W4312551440 hasConcept C120314980 @default.
- W4312551440 hasConcept C127413603 @default.
- W4312551440 hasConcept C130120984 @default.
- W4312551440 hasConcept C134306372 @default.
- W4312551440 hasConcept C162324750 @default.
- W4312551440 hasConcept C203616005 @default.
- W4312551440 hasConcept C27753989 @default.
- W4312551440 hasConcept C2777303404 @default.
- W4312551440 hasConcept C28826006 @default.
- W4312551440 hasConcept C33923547 @default.
- W4312551440 hasConcept C41008148 @default.
- W4312551440 hasConcept C50522688 @default.
- W4312551440 hasConcept C57273362 @default.
- W4312551440 hasConcept C62611344 @default.
- W4312551440 hasConcept C66938386 @default.
- W4312551440 hasConcept C72134830 @default.
- W4312551440 hasConcept C80444323 @default.
- W4312551440 hasConcept C81978471 @default.
- W4312551440 hasConceptScore W4312551440C106159729 @default.
- W4312551440 hasConceptScore W4312551440C11413529 @default.
- W4312551440 hasConceptScore W4312551440C120314980 @default.
- W4312551440 hasConceptScore W4312551440C127413603 @default.
- W4312551440 hasConceptScore W4312551440C130120984 @default.
- W4312551440 hasConceptScore W4312551440C134306372 @default.
- W4312551440 hasConceptScore W4312551440C162324750 @default.
- W4312551440 hasConceptScore W4312551440C203616005 @default.
- W4312551440 hasConceptScore W4312551440C27753989 @default.
- W4312551440 hasConceptScore W4312551440C2777303404 @default.
- W4312551440 hasConceptScore W4312551440C28826006 @default.
- W4312551440 hasConceptScore W4312551440C33923547 @default.
- W4312551440 hasConceptScore W4312551440C41008148 @default.
- W4312551440 hasConceptScore W4312551440C50522688 @default.
- W4312551440 hasConceptScore W4312551440C57273362 @default.
- W4312551440 hasConceptScore W4312551440C62611344 @default.
- W4312551440 hasConceptScore W4312551440C66938386 @default.
- W4312551440 hasConceptScore W4312551440C72134830 @default.
- W4312551440 hasConceptScore W4312551440C80444323 @default.
- W4312551440 hasConceptScore W4312551440C81978471 @default.
- W4312551440 hasFunder F4320320847 @default.
- W4312551440 hasLocation W43125514401 @default.
- W4312551440 hasOpenAccess W4312551440 @default.
- W4312551440 hasPrimaryLocation W43125514401 @default.
- W4312551440 hasRelatedWork W1482675070 @default.
- W4312551440 hasRelatedWork W1485627940 @default.
- W4312551440 hasRelatedWork W1819217 @default.
- W4312551440 hasRelatedWork W2000654528 @default.
- W4312551440 hasRelatedWork W2019952268 @default.
- W4312551440 hasRelatedWork W2280258626 @default.
- W4312551440 hasRelatedWork W2394192508 @default.
- W4312551440 hasRelatedWork W2790547488 @default.
- W4312551440 hasRelatedWork W2981239589 @default.
- W4312551440 hasRelatedWork W3027133497 @default.
- W4312551440 isParatext "false" @default.
- W4312551440 isRetracted "false" @default.
- W4312551440 workType "article" @default.