Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310998032> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4310998032 abstract "Many modern software systems are enabled by deep learning libraries such as TensorFlow and PyTorch. As deep learning is now prevalent, the security of deep learning libraries is a key concern. Fuzzing deep learning libraries presents two challenges. Firstly, to reach the functionality of the libraries, fuzzers have to use inputs from the valid input domain of each API function, which may be unknown. Secondly, many inputs are redundant. Randomly sampled invalid inputs are likely not to trigger new behaviors. While existing approaches partially address the first challenge, they overlook the second challenge. We propose SkipFuzz, an approach for fuzzing deep learning libraries. To generate valid inputs, SkipFuzz learns the input constraints of each API function using active learning. By using information gained during fuzzing, SkipFuzz infers a model of the input constraints, and, thus, generate valid inputs. SkipFuzz comprises an active learner which queries a test executor to obtain feedback for inference. After constructing hypotheses, the active learner poses queries and refines the hypotheses using the feedback from the test executor, which indicates if the library accepts or rejects an input, i.e., if it satisfies the input constraints or not. Inputs from different categories are used to invoke the library to check if a set of inputs satisfies a function's input constraints. Inputs in one category are distinguished from other categories by possible input constraints they would satisfy, e.g. they are tensors of a certain shape. As such, SkipFuzz is able to refine its hypothesis by eliminating possible candidates of the input constraints. This active learning-based approach addresses the challenge of redundant inputs. Using SkipFuzz, we have found and reported 43 crashes. 28 of them have been confirmed, with 13 unique CVEs assigned." @default.
- W4310998032 created "2022-12-22" @default.
- W4310998032 creator A5024592388 @default.
- W4310998032 creator A5027335548 @default.
- W4310998032 creator A5033078494 @default.
- W4310998032 creator A5053134485 @default.
- W4310998032 creator A5081036622 @default.
- W4310998032 creator A5081510237 @default.
- W4310998032 creator A5088206879 @default.
- W4310998032 date "2022-12-07" @default.
- W4310998032 modified "2023-10-17" @default.
- W4310998032 title "SkipFuzz: Active Learning-based Input Selection for Fuzzing Deep Learning Libraries" @default.
- W4310998032 doi "https://doi.org/10.48550/arxiv.2212.04038" @default.
- W4310998032 hasPublicationYear "2022" @default.
- W4310998032 type Work @default.
- W4310998032 citedByCount "0" @default.
- W4310998032 crossrefType "posted-content" @default.
- W4310998032 hasAuthorship W4310998032A5024592388 @default.
- W4310998032 hasAuthorship W4310998032A5027335548 @default.
- W4310998032 hasAuthorship W4310998032A5033078494 @default.
- W4310998032 hasAuthorship W4310998032A5053134485 @default.
- W4310998032 hasAuthorship W4310998032A5081036622 @default.
- W4310998032 hasAuthorship W4310998032A5081510237 @default.
- W4310998032 hasAuthorship W4310998032A5088206879 @default.
- W4310998032 hasBestOaLocation W43109980321 @default.
- W4310998032 hasConcept C108583219 @default.
- W4310998032 hasConcept C111065885 @default.
- W4310998032 hasConcept C119857082 @default.
- W4310998032 hasConcept C134306372 @default.
- W4310998032 hasConcept C14036430 @default.
- W4310998032 hasConcept C154945302 @default.
- W4310998032 hasConcept C177264268 @default.
- W4310998032 hasConcept C17744445 @default.
- W4310998032 hasConcept C180591056 @default.
- W4310998032 hasConcept C199360897 @default.
- W4310998032 hasConcept C199539241 @default.
- W4310998032 hasConcept C26517878 @default.
- W4310998032 hasConcept C2776214188 @default.
- W4310998032 hasConcept C2777904410 @default.
- W4310998032 hasConcept C33923547 @default.
- W4310998032 hasConcept C36503486 @default.
- W4310998032 hasConcept C38652104 @default.
- W4310998032 hasConcept C41008148 @default.
- W4310998032 hasConcept C78458016 @default.
- W4310998032 hasConcept C81917197 @default.
- W4310998032 hasConcept C86803240 @default.
- W4310998032 hasConceptScore W4310998032C108583219 @default.
- W4310998032 hasConceptScore W4310998032C111065885 @default.
- W4310998032 hasConceptScore W4310998032C119857082 @default.
- W4310998032 hasConceptScore W4310998032C134306372 @default.
- W4310998032 hasConceptScore W4310998032C14036430 @default.
- W4310998032 hasConceptScore W4310998032C154945302 @default.
- W4310998032 hasConceptScore W4310998032C177264268 @default.
- W4310998032 hasConceptScore W4310998032C17744445 @default.
- W4310998032 hasConceptScore W4310998032C180591056 @default.
- W4310998032 hasConceptScore W4310998032C199360897 @default.
- W4310998032 hasConceptScore W4310998032C199539241 @default.
- W4310998032 hasConceptScore W4310998032C26517878 @default.
- W4310998032 hasConceptScore W4310998032C2776214188 @default.
- W4310998032 hasConceptScore W4310998032C2777904410 @default.
- W4310998032 hasConceptScore W4310998032C33923547 @default.
- W4310998032 hasConceptScore W4310998032C36503486 @default.
- W4310998032 hasConceptScore W4310998032C38652104 @default.
- W4310998032 hasConceptScore W4310998032C41008148 @default.
- W4310998032 hasConceptScore W4310998032C78458016 @default.
- W4310998032 hasConceptScore W4310998032C81917197 @default.
- W4310998032 hasConceptScore W4310998032C86803240 @default.
- W4310998032 hasLocation W43109980321 @default.
- W4310998032 hasLocation W43109980322 @default.
- W4310998032 hasOpenAccess W4310998032 @default.
- W4310998032 hasPrimaryLocation W43109980321 @default.
- W4310998032 hasRelatedWork W2230606172 @default.
- W4310998032 hasRelatedWork W2511770387 @default.
- W4310998032 hasRelatedWork W3120811337 @default.
- W4310998032 hasRelatedWork W4237320244 @default.
- W4310998032 hasRelatedWork W3015007115 @default.
- W4310998032 hasRelatedWork W3016598040 @default.
- W4310998032 hasRelatedWork W3036516033 @default.
- W4310998032 hasRelatedWork W3044912482 @default.
- W4310998032 hasRelatedWork W3107299409 @default.
- W4310998032 hasRelatedWork W3132876088 @default.
- W4310998032 isParatext "false" @default.
- W4310998032 isRetracted "false" @default.
- W4310998032 workType "article" @default.