Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287111101> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W4287111101 abstract "Federated Learning (FL) is a promising decentralized learning framework and has great potentials in privacy preservation and in lowering the computation load at the cloud. Recent work showed that FedAvg and FedProx - the two widely-adopted FL algorithms - fail to reach the stationary points of the global optimization objective even for homogeneous linear regression problems. Further, it is concerned that the common model learned might not generalize well locally at all in the presence of heterogeneity. In this paper, we analyze the convergence and statistical efficiency of FedAvg and FedProx, addressing the above two concerns. Our analysis is based on the standard non-parametric regression in a reproducing kernel Hilbert space (RKHS), and allows for heterogeneous local data distributions and unbalanced local datasets. We prove that the estimation errors, measured in either the empirical norm or the RKHS norm, decay with a rate of 1/t in general and exponentially for finite-rank kernels. In certain heterogeneous settings, these upper bounds also imply that both FedAvg and FedProx achieve the optimal error rate. To further analytically quantify the impact of the heterogeneity at each client, we propose and characterize a novel notion-federation gain, defined as the reduction of the estimation error for a client to join the FL. We discover that when the data heterogeneity is moderate, a client with limited local data can benefit from a common model with a large federation gain. Numerical experiments further corroborate our theoretical findings." @default.
- W4287111101 created "2022-07-25" @default.
- W4287111101 creator A5042100539 @default.
- W4287111101 creator A5049410873 @default.
- W4287111101 creator A5050899081 @default.
- W4287111101 date "2021-06-29" @default.
- W4287111101 modified "2023-09-30" @default.
- W4287111101 title "A Non-parametric View of FedAvg and FedProx: Beyond Stationary Points" @default.
- W4287111101 doi "https://doi.org/10.48550/arxiv.2106.15216" @default.
- W4287111101 hasPublicationYear "2021" @default.
- W4287111101 type Work @default.
- W4287111101 citedByCount "1" @default.
- W4287111101 countsByYear W42871111012023 @default.
- W4287111101 crossrefType "posted-content" @default.
- W4287111101 hasAuthorship W4287111101A5042100539 @default.
- W4287111101 hasAuthorship W4287111101A5049410873 @default.
- W4287111101 hasAuthorship W4287111101A5050899081 @default.
- W4287111101 hasBestOaLocation W42871111011 @default.
- W4287111101 hasConcept C105795698 @default.
- W4287111101 hasConcept C11413529 @default.
- W4287111101 hasConcept C117251300 @default.
- W4287111101 hasConcept C118615104 @default.
- W4287111101 hasConcept C126255220 @default.
- W4287111101 hasConcept C127162648 @default.
- W4287111101 hasConcept C134306372 @default.
- W4287111101 hasConcept C162324750 @default.
- W4287111101 hasConcept C17744445 @default.
- W4287111101 hasConcept C191795146 @default.
- W4287111101 hasConcept C199539241 @default.
- W4287111101 hasConcept C2777303404 @default.
- W4287111101 hasConcept C28826006 @default.
- W4287111101 hasConcept C31258907 @default.
- W4287111101 hasConcept C33923547 @default.
- W4287111101 hasConcept C41008148 @default.
- W4287111101 hasConcept C45374587 @default.
- W4287111101 hasConcept C50522688 @default.
- W4287111101 hasConcept C57869625 @default.
- W4287111101 hasConcept C62799726 @default.
- W4287111101 hasConcept C74193536 @default.
- W4287111101 hasConcept C80884492 @default.
- W4287111101 hasConceptScore W4287111101C105795698 @default.
- W4287111101 hasConceptScore W4287111101C11413529 @default.
- W4287111101 hasConceptScore W4287111101C117251300 @default.
- W4287111101 hasConceptScore W4287111101C118615104 @default.
- W4287111101 hasConceptScore W4287111101C126255220 @default.
- W4287111101 hasConceptScore W4287111101C127162648 @default.
- W4287111101 hasConceptScore W4287111101C134306372 @default.
- W4287111101 hasConceptScore W4287111101C162324750 @default.
- W4287111101 hasConceptScore W4287111101C17744445 @default.
- W4287111101 hasConceptScore W4287111101C191795146 @default.
- W4287111101 hasConceptScore W4287111101C199539241 @default.
- W4287111101 hasConceptScore W4287111101C2777303404 @default.
- W4287111101 hasConceptScore W4287111101C28826006 @default.
- W4287111101 hasConceptScore W4287111101C31258907 @default.
- W4287111101 hasConceptScore W4287111101C33923547 @default.
- W4287111101 hasConceptScore W4287111101C41008148 @default.
- W4287111101 hasConceptScore W4287111101C45374587 @default.
- W4287111101 hasConceptScore W4287111101C50522688 @default.
- W4287111101 hasConceptScore W4287111101C57869625 @default.
- W4287111101 hasConceptScore W4287111101C62799726 @default.
- W4287111101 hasConceptScore W4287111101C74193536 @default.
- W4287111101 hasConceptScore W4287111101C80884492 @default.
- W4287111101 hasLocation W42871111011 @default.
- W4287111101 hasOpenAccess W4287111101 @default.
- W4287111101 hasPrimaryLocation W42871111011 @default.
- W4287111101 hasRelatedWork W1999906421 @default.
- W4287111101 hasRelatedWork W2035142076 @default.
- W4287111101 hasRelatedWork W2038766029 @default.
- W4287111101 hasRelatedWork W2055872486 @default.
- W4287111101 hasRelatedWork W2113250618 @default.
- W4287111101 hasRelatedWork W2188283173 @default.
- W4287111101 hasRelatedWork W2896790383 @default.
- W4287111101 hasRelatedWork W2949326632 @default.
- W4287111101 hasRelatedWork W4249432834 @default.
- W4287111101 hasRelatedWork W4286233510 @default.
- W4287111101 isParatext "false" @default.
- W4287111101 isRetracted "false" @default.
- W4287111101 workType "article" @default.