Matches in SemOpenAlex for { <https://semopenalex.org/work/W2895938368> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W2895938368 abstract "We give a new algorithm for learning a two-layer neural network under a general class of input distributions. Assuming there is a ground-truth two-layer network $$ y = A sigma(Wx) + xi, $$ where $A,W$ are weight matrices, $xi$ represents noise, and the number of neurons in the hidden layer is no larger than the input or output, our algorithm is guaranteed to recover the parameters $A,W$ of the ground-truth network. The only requirement on the input $x$ is that it is symmetric, which still allows highly complicated and structured input. Our algorithm is based on the method-of-moments framework and extends several results in tensor decompositions. We use spectral algorithms to avoid the complicated non-convex optimization in learning neural networks. Experiments show that our algorithm can robustly learn the ground-truth neural network with a small number of samples for many symmetric input distributions." @default.
- W2895938368 created "2018-10-26" @default.
- W2895938368 creator A5035001911 @default.
- W2895938368 creator A5044936528 @default.
- W2895938368 creator A5060340617 @default.
- W2895938368 creator A5080647450 @default.
- W2895938368 date "2018-09-27" @default.
- W2895938368 modified "2023-10-16" @default.
- W2895938368 title "Learning Two-Layer Neural Networks with Symmetric Inputs" @default.
- W2895938368 hasPublicationYear "2018" @default.
- W2895938368 type Work @default.
- W2895938368 sameAs 2895938368 @default.
- W2895938368 citedByCount "8" @default.
- W2895938368 countsByYear W28959383682019 @default.
- W2895938368 countsByYear W28959383682020 @default.
- W2895938368 countsByYear W28959383682021 @default.
- W2895938368 crossrefType "proceedings-article" @default.
- W2895938368 hasAuthorship W2895938368A5035001911 @default.
- W2895938368 hasAuthorship W2895938368A5044936528 @default.
- W2895938368 hasAuthorship W2895938368A5060340617 @default.
- W2895938368 hasAuthorship W2895938368A5080647450 @default.
- W2895938368 hasConcept C11413529 @default.
- W2895938368 hasConcept C115961682 @default.
- W2895938368 hasConcept C121332964 @default.
- W2895938368 hasConcept C146849305 @default.
- W2895938368 hasConcept C154945302 @default.
- W2895938368 hasConcept C155281189 @default.
- W2895938368 hasConcept C175202392 @default.
- W2895938368 hasConcept C178790620 @default.
- W2895938368 hasConcept C185592680 @default.
- W2895938368 hasConcept C202444582 @default.
- W2895938368 hasConcept C2778049214 @default.
- W2895938368 hasConcept C2779227376 @default.
- W2895938368 hasConcept C33923547 @default.
- W2895938368 hasConcept C41008148 @default.
- W2895938368 hasConcept C50644808 @default.
- W2895938368 hasConcept C62520636 @default.
- W2895938368 hasConcept C86582703 @default.
- W2895938368 hasConcept C99498987 @default.
- W2895938368 hasConceptScore W2895938368C11413529 @default.
- W2895938368 hasConceptScore W2895938368C115961682 @default.
- W2895938368 hasConceptScore W2895938368C121332964 @default.
- W2895938368 hasConceptScore W2895938368C146849305 @default.
- W2895938368 hasConceptScore W2895938368C154945302 @default.
- W2895938368 hasConceptScore W2895938368C155281189 @default.
- W2895938368 hasConceptScore W2895938368C175202392 @default.
- W2895938368 hasConceptScore W2895938368C178790620 @default.
- W2895938368 hasConceptScore W2895938368C185592680 @default.
- W2895938368 hasConceptScore W2895938368C202444582 @default.
- W2895938368 hasConceptScore W2895938368C2778049214 @default.
- W2895938368 hasConceptScore W2895938368C2779227376 @default.
- W2895938368 hasConceptScore W2895938368C33923547 @default.
- W2895938368 hasConceptScore W2895938368C41008148 @default.
- W2895938368 hasConceptScore W2895938368C50644808 @default.
- W2895938368 hasConceptScore W2895938368C62520636 @default.
- W2895938368 hasConceptScore W2895938368C86582703 @default.
- W2895938368 hasConceptScore W2895938368C99498987 @default.
- W2895938368 hasLocation W28959383681 @default.
- W2895938368 hasOpenAccess W2895938368 @default.
- W2895938368 hasPrimaryLocation W28959383681 @default.
- W2895938368 hasRelatedWork W1529647383 @default.
- W2895938368 hasRelatedWork W1680674731 @default.
- W2895938368 hasRelatedWork W2096632339 @default.
- W2895938368 hasRelatedWork W2172308869 @default.
- W2895938368 hasRelatedWork W2319967243 @default.
- W2895938368 hasRelatedWork W2558384550 @default.
- W2895938368 hasRelatedWork W2560504092 @default.
- W2895938368 hasRelatedWork W2899748887 @default.
- W2895938368 hasRelatedWork W2946510242 @default.
- W2895938368 hasRelatedWork W2949798199 @default.
- W2895938368 hasRelatedWork W2962767131 @default.
- W2895938368 hasRelatedWork W2965677819 @default.
- W2895938368 hasRelatedWork W2966376085 @default.
- W2895938368 hasRelatedWork W2984564059 @default.
- W2895938368 hasRelatedWork W3002939781 @default.
- W2895938368 hasRelatedWork W3095034705 @default.
- W2895938368 hasRelatedWork W3096184750 @default.
- W2895938368 hasRelatedWork W3133218296 @default.
- W2895938368 hasRelatedWork W3171826238 @default.
- W2895938368 hasRelatedWork W2187753657 @default.
- W2895938368 isParatext "false" @default.
- W2895938368 isRetracted "false" @default.
- W2895938368 magId "2895938368" @default.
- W2895938368 workType "article" @default.