Matches in SemOpenAlex for { <https://semopenalex.org/work/W2966854558> ?p ?o ?g. }
- W2966854558 abstract "We consider learning in decentralized heterogeneous networks: agents seek to minimize a convex functional that aggregates data across the network, while only having access to their local data streams. We focus on the case where agents seek to estimate a regression emph{function} that belongs to a reproducing kernel Hilbert space (RKHS). To incentivize coordination while respecting network heterogeneity, we impose nonlinear proximity constraints. To solve the constrained stochastic program, we propose applying a functional variant of stochastic primal-dual (Arrow-Hurwicz) method which yields a decentralized algorithm. To handle the fact that agents' functions have complexity proportional to time (owing to the RKHS parameterization), we project the primal iterates onto subspaces greedily constructed from kernel evaluations of agents' local observations. The resulting scheme, dubbed Heterogeneous Adaptive Learning with Kernels (HALK), when used with constant step-sizes, yields $mathcal{O}(sqrt{T})$ attenuation in sub-optimality and exactly satisfies the constraints in the long run, which improves upon the state of the art rates for vector-valued problems." @default.
- W2966854558 created "2019-08-13" @default.
- W2966854558 creator A5007246073 @default.
- W2966854558 creator A5025896653 @default.
- W2966854558 creator A5030276716 @default.
- W2966854558 creator A5039563144 @default.
- W2966854558 date "2019-08-01" @default.
- W2966854558 modified "2023-10-01" @default.
- W2966854558 title "Adaptive Kernel Learning in Heterogeneous Networks" @default.
- W2966854558 cites W1520045492 @default.
- W2966854558 cites W1540155273 @default.
- W2966854558 cites W1994616650 @default.
- W2966854558 cites W2015836240 @default.
- W2966854558 cites W2015904350 @default.
- W2966854558 cites W2020123437 @default.
- W2966854558 cites W2050643117 @default.
- W2966854558 cites W2066332749 @default.
- W2966854558 cites W2087558162 @default.
- W2966854558 cites W2095895508 @default.
- W2966854558 cites W2107396783 @default.
- W2966854558 cites W2116680122 @default.
- W2966854558 cites W2123705108 @default.
- W2966854558 cites W2124776405 @default.
- W2966854558 cites W2129809168 @default.
- W2966854558 cites W2150621701 @default.
- W2966854558 cites W2151693816 @default.
- W2966854558 cites W2153290280 @default.
- W2966854558 cites W2153663612 @default.
- W2966854558 cites W2155973058 @default.
- W2966854558 cites W2162195727 @default.
- W2966854558 cites W2405727896 @default.
- W2966854558 cites W2501301199 @default.
- W2966854558 cites W2516604945 @default.
- W2966854558 cites W2599228073 @default.
- W2966854558 cites W2763948036 @default.
- W2966854558 cites W2910292221 @default.
- W2966854558 cites W2912289066 @default.
- W2966854558 cites W2912682655 @default.
- W2966854558 cites W2919539866 @default.
- W2966854558 cites W2920025365 @default.
- W2966854558 cites W2928539936 @default.
- W2966854558 cites W2963070834 @default.
- W2966854558 cites W2963195786 @default.
- W2966854558 cites W2964252810 @default.
- W2966854558 cites W2966417857 @default.
- W2966854558 cites W3004065524 @default.
- W2966854558 cites W3008594870 @default.
- W2966854558 cites W3034445294 @default.
- W2966854558 cites W3124367822 @default.
- W2966854558 cites W88657458 @default.
- W2966854558 cites W996830848 @default.
- W2966854558 hasPublicationYear "2019" @default.
- W2966854558 type Work @default.
- W2966854558 sameAs 2966854558 @default.
- W2966854558 citedByCount "0" @default.
- W2966854558 crossrefType "posted-content" @default.
- W2966854558 hasAuthorship W2966854558A5007246073 @default.
- W2966854558 hasAuthorship W2966854558A5025896653 @default.
- W2966854558 hasAuthorship W2966854558A5030276716 @default.
- W2966854558 hasAuthorship W2966854558A5039563144 @default.
- W2966854558 hasConcept C118615104 @default.
- W2966854558 hasConcept C120665830 @default.
- W2966854558 hasConcept C121332964 @default.
- W2966854558 hasConcept C122280245 @default.
- W2966854558 hasConcept C12267149 @default.
- W2966854558 hasConcept C12362212 @default.
- W2966854558 hasConcept C126255220 @default.
- W2966854558 hasConcept C134306372 @default.
- W2966854558 hasConcept C140479938 @default.
- W2966854558 hasConcept C154945302 @default.
- W2966854558 hasConcept C158968445 @default.
- W2966854558 hasConcept C192209626 @default.
- W2966854558 hasConcept C202444582 @default.
- W2966854558 hasConcept C33923547 @default.
- W2966854558 hasConcept C41008148 @default.
- W2966854558 hasConcept C62799726 @default.
- W2966854558 hasConcept C74193536 @default.
- W2966854558 hasConcept C80884492 @default.
- W2966854558 hasConceptScore W2966854558C118615104 @default.
- W2966854558 hasConceptScore W2966854558C120665830 @default.
- W2966854558 hasConceptScore W2966854558C121332964 @default.
- W2966854558 hasConceptScore W2966854558C122280245 @default.
- W2966854558 hasConceptScore W2966854558C12267149 @default.
- W2966854558 hasConceptScore W2966854558C12362212 @default.
- W2966854558 hasConceptScore W2966854558C126255220 @default.
- W2966854558 hasConceptScore W2966854558C134306372 @default.
- W2966854558 hasConceptScore W2966854558C140479938 @default.
- W2966854558 hasConceptScore W2966854558C154945302 @default.
- W2966854558 hasConceptScore W2966854558C158968445 @default.
- W2966854558 hasConceptScore W2966854558C192209626 @default.
- W2966854558 hasConceptScore W2966854558C202444582 @default.
- W2966854558 hasConceptScore W2966854558C33923547 @default.
- W2966854558 hasConceptScore W2966854558C41008148 @default.
- W2966854558 hasConceptScore W2966854558C62799726 @default.
- W2966854558 hasConceptScore W2966854558C74193536 @default.
- W2966854558 hasConceptScore W2966854558C80884492 @default.
- W2966854558 hasLocation W29668545581 @default.
- W2966854558 hasOpenAccess W2966854558 @default.
- W2966854558 hasPrimaryLocation W29668545581 @default.
- W2966854558 hasRelatedWork W2012106138 @default.