Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225756064> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W4225756064 abstract "Data poisoning for reinforcement learning has historically focused on general performance degradation, and targeted attacks have been successful via perturbations that involve control of the victim's policy and rewards. We introduce an insidious poisoning attack for reinforcement learning which causes agent misbehavior only at specific target states - all while minimally modifying a small fraction of training observations without assuming any control over policy or reward. We accomplish this by adapting a recent technique, gradient alignment, to reinforcement learning. We test our method and demonstrate success in two Atari games of varying difficulty." @default.
- W4225756064 created "2022-05-05" @default.
- W4225756064 creator A5034473032 @default.
- W4225756064 creator A5043788981 @default.
- W4225756064 creator A5060687985 @default.
- W4225756064 creator A5065987967 @default.
- W4225756064 date "2022-01-03" @default.
- W4225756064 modified "2023-09-25" @default.
- W4225756064 title "Execute Order 66: Targeted Data Poisoning for Reinforcement Learning" @default.
- W4225756064 doi "https://doi.org/10.48550/arxiv.2201.00762" @default.
- W4225756064 hasPublicationYear "2022" @default.
- W4225756064 type Work @default.
- W4225756064 citedByCount "0" @default.
- W4225756064 crossrefType "posted-content" @default.
- W4225756064 hasAuthorship W4225756064A5034473032 @default.
- W4225756064 hasAuthorship W4225756064A5043788981 @default.
- W4225756064 hasAuthorship W4225756064A5060687985 @default.
- W4225756064 hasAuthorship W4225756064A5065987967 @default.
- W4225756064 hasBestOaLocation W42257560641 @default.
- W4225756064 hasConcept C10138342 @default.
- W4225756064 hasConcept C144133560 @default.
- W4225756064 hasConcept C154945302 @default.
- W4225756064 hasConcept C15744967 @default.
- W4225756064 hasConcept C182306322 @default.
- W4225756064 hasConcept C2775924081 @default.
- W4225756064 hasConcept C38652104 @default.
- W4225756064 hasConcept C41008148 @default.
- W4225756064 hasConcept C67203356 @default.
- W4225756064 hasConcept C77805123 @default.
- W4225756064 hasConcept C97541855 @default.
- W4225756064 hasConceptScore W4225756064C10138342 @default.
- W4225756064 hasConceptScore W4225756064C144133560 @default.
- W4225756064 hasConceptScore W4225756064C154945302 @default.
- W4225756064 hasConceptScore W4225756064C15744967 @default.
- W4225756064 hasConceptScore W4225756064C182306322 @default.
- W4225756064 hasConceptScore W4225756064C2775924081 @default.
- W4225756064 hasConceptScore W4225756064C38652104 @default.
- W4225756064 hasConceptScore W4225756064C41008148 @default.
- W4225756064 hasConceptScore W4225756064C67203356 @default.
- W4225756064 hasConceptScore W4225756064C77805123 @default.
- W4225756064 hasConceptScore W4225756064C97541855 @default.
- W4225756064 hasLocation W42257560641 @default.
- W4225756064 hasOpenAccess W4225756064 @default.
- W4225756064 hasPrimaryLocation W42257560641 @default.
- W4225756064 hasRelatedWork W1323832 @default.
- W4225756064 hasRelatedWork W13374848 @default.
- W4225756064 hasRelatedWork W2683128 @default.
- W4225756064 hasRelatedWork W3471107 @default.
- W4225756064 hasRelatedWork W4651166 @default.
- W4225756064 hasRelatedWork W5081013 @default.
- W4225756064 hasRelatedWork W5435649 @default.
- W4225756064 hasRelatedWork W5779190 @default.
- W4225756064 hasRelatedWork W8539471 @default.
- W4225756064 hasRelatedWork W8637261 @default.
- W4225756064 isParatext "false" @default.
- W4225756064 isRetracted "false" @default.
- W4225756064 workType "article" @default.