Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312806467> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4312806467 abstract "Massive datasets are typically distributed geographically across multiple sites, where scalability, data privacy and integrity, as well as bandwidth scarcity typically discourage uploading these data to a central server. This has propelled the so-called federated learning framework where multiple workers exchange information with a server to learn a “centralized” model using data locally generated and/or stored across workers. This learning framework necessitates workers to communicate iteratively with the server. Although appealing for its scalability, one needs to carefully address the various data distribution shifts across workers, which degrades the performance of the learnt model. In this context, the distributionally robust op-timization framework is considered here. The objective is to endow the trained model with robustness against adversarially manipulated input data, or, distributional uncertainties, such as mismatches between training and testing data distributions, or among datasets stored at different workers. To this aim, the data distribution is assumed unknown, and to land within a Wasserstein ball centered around the empirical data distribution. This robust learning task entails an infinite-dimensional optimization problem, which is challenging. Leveraging a strong duality result, a surrogate is obtained, for which a primal-dual algorithm is developed. Compared to classical methods, the proposed algorithm offers robustness with little computational overhead. Numerical tests using image datasets showcase the merits of the proposed algorithm under several existing adversarial attacks and distributional uncertainties." @default.
- W4312806467 created "2023-01-05" @default.
- W4312806467 creator A5016514790 @default.
- W4312806467 creator A5026758314 @default.
- W4312806467 date "2022-08-29" @default.
- W4312806467 modified "2023-09-27" @default.
- W4312806467 title "Learning while Respecting Privacy and Robustness to Adversarial Distributed Datasets" @default.
- W4312806467 cites W1968355947 @default.
- W4312806467 cites W2048028546 @default.
- W4312806467 cites W2125417745 @default.
- W4312806467 cites W2535873859 @default.
- W4312806467 cites W2885208219 @default.
- W4312806467 cites W2924551358 @default.
- W4312806467 cites W3010823055 @default.
- W4312806467 cites W3015636663 @default.
- W4312806467 cites W3103802018 @default.
- W4312806467 cites W3209083247 @default.
- W4312806467 doi "https://doi.org/10.23919/eusipco55093.2022.9909977" @default.
- W4312806467 hasPublicationYear "2022" @default.
- W4312806467 type Work @default.
- W4312806467 citedByCount "0" @default.
- W4312806467 crossrefType "proceedings-article" @default.
- W4312806467 hasAuthorship W4312806467A5016514790 @default.
- W4312806467 hasAuthorship W4312806467A5026758314 @default.
- W4312806467 hasConcept C104317684 @default.
- W4312806467 hasConcept C111919701 @default.
- W4312806467 hasConcept C119857082 @default.
- W4312806467 hasConcept C120314980 @default.
- W4312806467 hasConcept C124101348 @default.
- W4312806467 hasConcept C154945302 @default.
- W4312806467 hasConcept C185592680 @default.
- W4312806467 hasConcept C37736160 @default.
- W4312806467 hasConcept C41008148 @default.
- W4312806467 hasConcept C48044578 @default.
- W4312806467 hasConcept C55493867 @default.
- W4312806467 hasConcept C63479239 @default.
- W4312806467 hasConcept C67186912 @default.
- W4312806467 hasConcept C71901391 @default.
- W4312806467 hasConcept C77088390 @default.
- W4312806467 hasConceptScore W4312806467C104317684 @default.
- W4312806467 hasConceptScore W4312806467C111919701 @default.
- W4312806467 hasConceptScore W4312806467C119857082 @default.
- W4312806467 hasConceptScore W4312806467C120314980 @default.
- W4312806467 hasConceptScore W4312806467C124101348 @default.
- W4312806467 hasConceptScore W4312806467C154945302 @default.
- W4312806467 hasConceptScore W4312806467C185592680 @default.
- W4312806467 hasConceptScore W4312806467C37736160 @default.
- W4312806467 hasConceptScore W4312806467C41008148 @default.
- W4312806467 hasConceptScore W4312806467C48044578 @default.
- W4312806467 hasConceptScore W4312806467C55493867 @default.
- W4312806467 hasConceptScore W4312806467C63479239 @default.
- W4312806467 hasConceptScore W4312806467C67186912 @default.
- W4312806467 hasConceptScore W4312806467C71901391 @default.
- W4312806467 hasConceptScore W4312806467C77088390 @default.
- W4312806467 hasLocation W43128064671 @default.
- W4312806467 hasOpenAccess W4312806467 @default.
- W4312806467 hasPrimaryLocation W43128064671 @default.
- W4312806467 hasRelatedWork W1596201972 @default.
- W4312806467 hasRelatedWork W1967954938 @default.
- W4312806467 hasRelatedWork W1986253068 @default.
- W4312806467 hasRelatedWork W2364921833 @default.
- W4312806467 hasRelatedWork W2380023786 @default.
- W4312806467 hasRelatedWork W2385146268 @default.
- W4312806467 hasRelatedWork W2385763152 @default.
- W4312806467 hasRelatedWork W2389719923 @default.
- W4312806467 hasRelatedWork W2546696010 @default.
- W4312806467 hasRelatedWork W4311734044 @default.
- W4312806467 isParatext "false" @default.
- W4312806467 isRetracted "false" @default.
- W4312806467 workType "article" @default.