Matches in SemOpenAlex for { <https://semopenalex.org/work/W3044696307> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W3044696307 abstract "We present a novel method for generating symbolic adversarial examples: input regions guaranteed to only contain adversarial examples for the given neural network. These regions can generate real-world adversarial examples as they summarize trillions of adversarial examples. We theoretically show that computing optimal symbolic adversarial examples is computationally expensive. We present a method for approximating optimal examples in a scalable manner. Our method first selectively uses adversarial attacks to generate a candidate region and then prunes this region with hyperplanes that fit points obtained via specialized sampling. It iterates until arriving at a symbolic adversarial example for which it can prove, via state-of-the-art convex relaxation techniques, that the region only contains adversarial examples. Our experimental results demonstrate that our method is practically effective: it only needs a few thousand attacks to infer symbolic summaries guaranteed to contain $approx 10^{258}$ adversarial examples." @default.
- W3044696307 created "2020-07-29" @default.
- W3044696307 creator A5002442193 @default.
- W3044696307 creator A5014657381 @default.
- W3044696307 creator A5045044745 @default.
- W3044696307 creator A5069901599 @default.
- W3044696307 date "2020-07-23" @default.
- W3044696307 modified "2023-09-27" @default.
- W3044696307 title "Scalable Inference of Symbolic Adversarial Examples." @default.
- W3044696307 cites W2112796928 @default.
- W3044696307 cites W2460937040 @default.
- W3044696307 cites W2590523583 @default.
- W3044696307 cites W2611576673 @default.
- W3044696307 cites W2807040120 @default.
- W3044696307 cites W2810611310 @default.
- W3044696307 cites W2899434523 @default.
- W3044696307 cites W2903571520 @default.
- W3044696307 cites W2948854542 @default.
- W3044696307 cites W2949389295 @default.
- W3044696307 cites W2962941306 @default.
- W3044696307 cites W2963207607 @default.
- W3044696307 cites W2963857521 @default.
- W3044696307 cites W2964153729 @default.
- W3044696307 cites W2981016037 @default.
- W3044696307 cites W2997223761 @default.
- W3044696307 cites W2999818223 @default.
- W3044696307 cites W3005260693 @default.
- W3044696307 cites W3035345420 @default.
- W3044696307 cites W3040452189 @default.
- W3044696307 cites W3103340107 @default.
- W3044696307 cites W3107235539 @default.
- W3044696307 cites W3118608800 @default.
- W3044696307 hasPublicationYear "2020" @default.
- W3044696307 type Work @default.
- W3044696307 sameAs 3044696307 @default.
- W3044696307 citedByCount "0" @default.
- W3044696307 crossrefType "posted-content" @default.
- W3044696307 hasAuthorship W3044696307A5002442193 @default.
- W3044696307 hasAuthorship W3044696307A5014657381 @default.
- W3044696307 hasAuthorship W3044696307A5045044745 @default.
- W3044696307 hasAuthorship W3044696307A5069901599 @default.
- W3044696307 hasConcept C11413529 @default.
- W3044696307 hasConcept C134306372 @default.
- W3044696307 hasConcept C140479938 @default.
- W3044696307 hasConcept C154945302 @default.
- W3044696307 hasConcept C15744967 @default.
- W3044696307 hasConcept C2524010 @default.
- W3044696307 hasConcept C2776029896 @default.
- W3044696307 hasConcept C2776214188 @default.
- W3044696307 hasConcept C33923547 @default.
- W3044696307 hasConcept C37736160 @default.
- W3044696307 hasConcept C41008148 @default.
- W3044696307 hasConcept C48044578 @default.
- W3044696307 hasConcept C68693459 @default.
- W3044696307 hasConcept C77088390 @default.
- W3044696307 hasConcept C77805123 @default.
- W3044696307 hasConcept C80444323 @default.
- W3044696307 hasConceptScore W3044696307C11413529 @default.
- W3044696307 hasConceptScore W3044696307C134306372 @default.
- W3044696307 hasConceptScore W3044696307C140479938 @default.
- W3044696307 hasConceptScore W3044696307C154945302 @default.
- W3044696307 hasConceptScore W3044696307C15744967 @default.
- W3044696307 hasConceptScore W3044696307C2524010 @default.
- W3044696307 hasConceptScore W3044696307C2776029896 @default.
- W3044696307 hasConceptScore W3044696307C2776214188 @default.
- W3044696307 hasConceptScore W3044696307C33923547 @default.
- W3044696307 hasConceptScore W3044696307C37736160 @default.
- W3044696307 hasConceptScore W3044696307C41008148 @default.
- W3044696307 hasConceptScore W3044696307C48044578 @default.
- W3044696307 hasConceptScore W3044696307C68693459 @default.
- W3044696307 hasConceptScore W3044696307C77088390 @default.
- W3044696307 hasConceptScore W3044696307C77805123 @default.
- W3044696307 hasConceptScore W3044696307C80444323 @default.
- W3044696307 hasLocation W30446963071 @default.
- W3044696307 hasOpenAccess W3044696307 @default.
- W3044696307 hasPrimaryLocation W30446963071 @default.
- W3044696307 hasRelatedWork W1506934790 @default.
- W3044696307 hasRelatedWork W1830383289 @default.
- W3044696307 hasRelatedWork W2141373061 @default.
- W3044696307 hasRelatedWork W2149652368 @default.
- W3044696307 hasRelatedWork W2513258002 @default.
- W3044696307 hasRelatedWork W2519484726 @default.
- W3044696307 hasRelatedWork W2782527665 @default.
- W3044696307 hasRelatedWork W2804224372 @default.
- W3044696307 hasRelatedWork W2884837949 @default.
- W3044696307 hasRelatedWork W2901015811 @default.
- W3044696307 hasRelatedWork W2964016574 @default.
- W3044696307 hasRelatedWork W2990365857 @default.
- W3044696307 hasRelatedWork W3094066269 @default.
- W3044696307 hasRelatedWork W3097031579 @default.
- W3044696307 hasRelatedWork W3105448663 @default.
- W3044696307 hasRelatedWork W3122957072 @default.
- W3044696307 hasRelatedWork W3128605735 @default.
- W3044696307 hasRelatedWork W3133739219 @default.
- W3044696307 hasRelatedWork W2148723623 @default.
- W3044696307 hasRelatedWork W2605063165 @default.
- W3044696307 isParatext "false" @default.
- W3044696307 isRetracted "false" @default.
- W3044696307 magId "3044696307" @default.
- W3044696307 workType "article" @default.