Matches in SemOpenAlex for { <https://semopenalex.org/work/W3089751199> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3089751199 abstract "Federated learning is a setting where agents, each with access to their own data source, combine models from local data to create a global model. If agents are drawing their data from different distributions, though, federated learning might produce a biased global model that is not optimal for each agent. This means that agents face a fundamental question: should they choose the global model or their local model? We show how this situation can be naturally analyzed through the framework of coalitional game theory. We propose the following game: there are heterogeneous players with different model parameters governing their data distribution and different amounts of data they have noisily drawn from their own distribution. Each player's goal is to obtain a model with minimal expected mean squared error (MSE) on their own distribution. They have a choice of fitting a model based solely on their own data, or combining their learned parameters with those of some subset of the other players. Combining models reduces the variance component of their error through access to more data, but increases the bias because of the heterogeneity of distributions. Here, we derive exact expected MSE values for problems in linear regression and mean estimation. We then analyze the resulting game in the framework of hedonic game theory; we study how players might divide into coalitions, where each set of players within a coalition jointly construct model(s). We analyze three methods of federation, modeling differing degrees of customization. In uniform federation, the agents collectively produce a single model. In coarse-grained federation, each agent can weight the global model together with their local model. In fine-grained federation, each agent can flexibly combine models from all other agents in the federation. For each method, we analyze the stable partitions of players into coalitions." @default.
- W3089751199 created "2020-10-08" @default.
- W3089751199 creator A5002338764 @default.
- W3089751199 creator A5055710645 @default.
- W3089751199 date "2020-10-02" @default.
- W3089751199 modified "2023-09-27" @default.
- W3089751199 title "Model-sharing Games: Analyzing Federated Learning Under Voluntary Participation" @default.
- W3089751199 cites W1991769471 @default.
- W3089751199 cites W2055862894 @default.
- W3089751199 cites W2063698478 @default.
- W3089751199 cites W2064054477 @default.
- W3089751199 cites W2134660637 @default.
- W3089751199 cites W2179028626 @default.
- W3089751199 cites W2752504160 @default.
- W3089751199 cites W2946193510 @default.
- W3089751199 cites W2995022099 @default.
- W3089751199 cites W3006157707 @default.
- W3089751199 cites W3006317386 @default.
- W3089751199 cites W3007548213 @default.
- W3089751199 cites W3012968339 @default.
- W3089751199 cites W3034879135 @default.
- W3089751199 cites W3042621011 @default.
- W3089751199 cites W3080934299 @default.
- W3089751199 cites W3081178085 @default.
- W3089751199 cites W3094458938 @default.
- W3089751199 cites W414222185 @default.
- W3089751199 hasPublicationYear "2020" @default.
- W3089751199 type Work @default.
- W3089751199 sameAs 3089751199 @default.
- W3089751199 citedByCount "1" @default.
- W3089751199 countsByYear W30897511992021 @default.
- W3089751199 crossrefType "posted-content" @default.
- W3089751199 hasAuthorship W3089751199A5002338764 @default.
- W3089751199 hasAuthorship W3089751199A5055710645 @default.
- W3089751199 hasConcept C105795698 @default.
- W3089751199 hasConcept C110121322 @default.
- W3089751199 hasConcept C121955636 @default.
- W3089751199 hasConcept C126255220 @default.
- W3089751199 hasConcept C134306372 @default.
- W3089751199 hasConcept C136764020 @default.
- W3089751199 hasConcept C139945424 @default.
- W3089751199 hasConcept C144133560 @default.
- W3089751199 hasConcept C154945302 @default.
- W3089751199 hasConcept C177264268 @default.
- W3089751199 hasConcept C183003079 @default.
- W3089751199 hasConcept C196083921 @default.
- W3089751199 hasConcept C199360897 @default.
- W3089751199 hasConcept C33923547 @default.
- W3089751199 hasConcept C41008148 @default.
- W3089751199 hasConceptScore W3089751199C105795698 @default.
- W3089751199 hasConceptScore W3089751199C110121322 @default.
- W3089751199 hasConceptScore W3089751199C121955636 @default.
- W3089751199 hasConceptScore W3089751199C126255220 @default.
- W3089751199 hasConceptScore W3089751199C134306372 @default.
- W3089751199 hasConceptScore W3089751199C136764020 @default.
- W3089751199 hasConceptScore W3089751199C139945424 @default.
- W3089751199 hasConceptScore W3089751199C144133560 @default.
- W3089751199 hasConceptScore W3089751199C154945302 @default.
- W3089751199 hasConceptScore W3089751199C177264268 @default.
- W3089751199 hasConceptScore W3089751199C183003079 @default.
- W3089751199 hasConceptScore W3089751199C196083921 @default.
- W3089751199 hasConceptScore W3089751199C199360897 @default.
- W3089751199 hasConceptScore W3089751199C33923547 @default.
- W3089751199 hasConceptScore W3089751199C41008148 @default.
- W3089751199 hasLocation W30897511991 @default.
- W3089751199 hasOpenAccess W3089751199 @default.
- W3089751199 hasPrimaryLocation W30897511991 @default.
- W3089751199 hasRelatedWork W1606611000 @default.
- W3089751199 hasRelatedWork W166697673 @default.
- W3089751199 hasRelatedWork W1889400262 @default.
- W3089751199 hasRelatedWork W1951940478 @default.
- W3089751199 hasRelatedWork W1997594734 @default.
- W3089751199 hasRelatedWork W2017483915 @default.
- W3089751199 hasRelatedWork W2050389111 @default.
- W3089751199 hasRelatedWork W2077202169 @default.
- W3089751199 hasRelatedWork W2107209546 @default.
- W3089751199 hasRelatedWork W2130056990 @default.
- W3089751199 hasRelatedWork W2134345668 @default.
- W3089751199 hasRelatedWork W2295312198 @default.
- W3089751199 hasRelatedWork W2787921040 @default.
- W3089751199 hasRelatedWork W2795787303 @default.
- W3089751199 hasRelatedWork W2947697349 @default.
- W3089751199 hasRelatedWork W2951258838 @default.
- W3089751199 hasRelatedWork W2963764295 @default.
- W3089751199 hasRelatedWork W2964141905 @default.
- W3089751199 hasRelatedWork W3173537849 @default.
- W3089751199 hasRelatedWork W3180384258 @default.
- W3089751199 isParatext "false" @default.
- W3089751199 isRetracted "false" @default.
- W3089751199 magId "3089751199" @default.
- W3089751199 workType "article" @default.