Matches in SemOpenAlex for { <https://semopenalex.org/work/W2967175067> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W2967175067 abstract "Analysis and manipulation of trained neural networks is a challenging and important problem. We propose a symbolic representation for piecewise-linear neural networks and discuss its efficient computation. With this representation, one can translate the problem of analyzing a complex neural network into that of analyzing a finite set of affine functions. We demonstrate the use of this representation for three applications. First, we apply the symbolic representation to computing weakest preconditions on network inputs, which we use to exactly visualize the advisories made by a network meant to operate an aircraft collision avoidance system. Second, we use the symbolic representation to compute strongest postconditions on the network outputs, which we use to perform bounded model checking on standard neural network controllers. Finally, we show how the symbolic representation can be combined with a new form of neural network to perform patching; i.e., correct user-specified behavior of the network." @default.
- W2967175067 created "2019-08-22" @default.
- W2967175067 creator A5029756316 @default.
- W2967175067 creator A5079165150 @default.
- W2967175067 date "2019-08-17" @default.
- W2967175067 modified "2023-09-24" @default.
- W2967175067 title "A Symbolic Neural Network Representation and its Application to Understanding, Verifying, and Patching Networks." @default.
- W2967175067 cites W2777417342 @default.
- W2967175067 hasPublicationYear "2019" @default.
- W2967175067 type Work @default.
- W2967175067 sameAs 2967175067 @default.
- W2967175067 citedByCount "0" @default.
- W2967175067 crossrefType "posted-content" @default.
- W2967175067 hasAuthorship W2967175067A5029756316 @default.
- W2967175067 hasAuthorship W2967175067A5079165150 @default.
- W2967175067 hasConcept C110251889 @default.
- W2967175067 hasConcept C134306372 @default.
- W2967175067 hasConcept C154945302 @default.
- W2967175067 hasConcept C177264268 @default.
- W2967175067 hasConcept C17744445 @default.
- W2967175067 hasConcept C199360897 @default.
- W2967175067 hasConcept C199539241 @default.
- W2967175067 hasConcept C202444582 @default.
- W2967175067 hasConcept C23123167 @default.
- W2967175067 hasConcept C2776359362 @default.
- W2967175067 hasConcept C33923547 @default.
- W2967175067 hasConcept C34388435 @default.
- W2967175067 hasConcept C41008148 @default.
- W2967175067 hasConcept C50644808 @default.
- W2967175067 hasConcept C65620979 @default.
- W2967175067 hasConcept C80444323 @default.
- W2967175067 hasConcept C92757383 @default.
- W2967175067 hasConcept C94625758 @default.
- W2967175067 hasConceptScore W2967175067C110251889 @default.
- W2967175067 hasConceptScore W2967175067C134306372 @default.
- W2967175067 hasConceptScore W2967175067C154945302 @default.
- W2967175067 hasConceptScore W2967175067C177264268 @default.
- W2967175067 hasConceptScore W2967175067C17744445 @default.
- W2967175067 hasConceptScore W2967175067C199360897 @default.
- W2967175067 hasConceptScore W2967175067C199539241 @default.
- W2967175067 hasConceptScore W2967175067C202444582 @default.
- W2967175067 hasConceptScore W2967175067C23123167 @default.
- W2967175067 hasConceptScore W2967175067C2776359362 @default.
- W2967175067 hasConceptScore W2967175067C33923547 @default.
- W2967175067 hasConceptScore W2967175067C34388435 @default.
- W2967175067 hasConceptScore W2967175067C41008148 @default.
- W2967175067 hasConceptScore W2967175067C50644808 @default.
- W2967175067 hasConceptScore W2967175067C65620979 @default.
- W2967175067 hasConceptScore W2967175067C80444323 @default.
- W2967175067 hasConceptScore W2967175067C92757383 @default.
- W2967175067 hasConceptScore W2967175067C94625758 @default.
- W2967175067 hasLocation W29671750671 @default.
- W2967175067 hasOpenAccess W2967175067 @default.
- W2967175067 hasPrimaryLocation W29671750671 @default.
- W2967175067 hasRelatedWork W1845899572 @default.
- W2967175067 hasRelatedWork W2065204741 @default.
- W2967175067 hasRelatedWork W2082959097 @default.
- W2967175067 hasRelatedWork W2158476197 @default.
- W2967175067 hasRelatedWork W2235915642 @default.
- W2967175067 hasRelatedWork W2292383450 @default.
- W2967175067 hasRelatedWork W2508259276 @default.
- W2967175067 hasRelatedWork W2559911651 @default.
- W2967175067 hasRelatedWork W2757956526 @default.
- W2967175067 hasRelatedWork W278347392 @default.
- W2967175067 hasRelatedWork W2962841959 @default.
- W2967175067 hasRelatedWork W2980740513 @default.
- W2967175067 hasRelatedWork W2996029397 @default.
- W2967175067 hasRelatedWork W3005885357 @default.
- W2967175067 hasRelatedWork W3131107316 @default.
- W2967175067 hasRelatedWork W3147413282 @default.
- W2967175067 hasRelatedWork W3182512154 @default.
- W2967175067 hasRelatedWork W2040364046 @default.
- W2967175067 hasRelatedWork W2182275691 @default.
- W2967175067 hasRelatedWork W3147872551 @default.
- W2967175067 isParatext "false" @default.
- W2967175067 isRetracted "false" @default.
- W2967175067 magId "2967175067" @default.
- W2967175067 workType "article" @default.