Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226227954> ?p ?o ?g. }
Showing items 1 to 64 of
64
with 100 items per page.
- W4226227954 abstract "Deep reinforcement learning is widely used to train autonomous cars in a simulated environment. Still, autonomous cars are well known for being vulnerable when exposed to adversarial attacks. This raises the question of whether we can train the adversary as a driving agent for finding failure scenarios in autonomous cars, and then retrain autonomous cars with new adversarial inputs to improve their robustness. In this work, we first train and compare adversarial car policy on two custom reward functions to test the driving control decision of autonomous cars in a multi-agent setting. Second, we verify that adversarial examples can be used not only for finding unwanted autonomous driving behavior, but also for helping autonomous driving cars in improving their deep reinforcement learning policies. By using a high fidelity urban driving simulation environment and vision-based driving agents, we demonstrate that the autonomous cars retrained using the adversary player noticeably increase the performance of their driving policies in terms of reducing collision and offroad steering errors." @default.
- W4226227954 created "2022-05-05" @default.
- W4226227954 creator A5056500610 @default.
- W4226227954 creator A5083305617 @default.
- W4226227954 date "2021-12-22" @default.
- W4226227954 modified "2023-09-29" @default.
- W4226227954 title "Adversarial Deep Reinforcement Learning for Trustworthy Autonomous Driving Policies" @default.
- W4226227954 hasPublicationYear "2021" @default.
- W4226227954 type Work @default.
- W4226227954 citedByCount "0" @default.
- W4226227954 crossrefType "posted-content" @default.
- W4226227954 hasAuthorship W4226227954A5056500610 @default.
- W4226227954 hasAuthorship W4226227954A5083305617 @default.
- W4226227954 hasBestOaLocation W42262279541 @default.
- W4226227954 hasConcept C104317684 @default.
- W4226227954 hasConcept C108583219 @default.
- W4226227954 hasConcept C121704057 @default.
- W4226227954 hasConcept C13687954 @default.
- W4226227954 hasConcept C153701036 @default.
- W4226227954 hasConcept C154945302 @default.
- W4226227954 hasConcept C185592680 @default.
- W4226227954 hasConcept C2776459999 @default.
- W4226227954 hasConcept C2780864053 @default.
- W4226227954 hasConcept C37736160 @default.
- W4226227954 hasConcept C38652104 @default.
- W4226227954 hasConcept C41008148 @default.
- W4226227954 hasConcept C41065033 @default.
- W4226227954 hasConcept C55493867 @default.
- W4226227954 hasConcept C63479239 @default.
- W4226227954 hasConcept C76155785 @default.
- W4226227954 hasConcept C97541855 @default.
- W4226227954 hasConceptScore W4226227954C104317684 @default.
- W4226227954 hasConceptScore W4226227954C108583219 @default.
- W4226227954 hasConceptScore W4226227954C121704057 @default.
- W4226227954 hasConceptScore W4226227954C13687954 @default.
- W4226227954 hasConceptScore W4226227954C153701036 @default.
- W4226227954 hasConceptScore W4226227954C154945302 @default.
- W4226227954 hasConceptScore W4226227954C185592680 @default.
- W4226227954 hasConceptScore W4226227954C2776459999 @default.
- W4226227954 hasConceptScore W4226227954C2780864053 @default.
- W4226227954 hasConceptScore W4226227954C37736160 @default.
- W4226227954 hasConceptScore W4226227954C38652104 @default.
- W4226227954 hasConceptScore W4226227954C41008148 @default.
- W4226227954 hasConceptScore W4226227954C41065033 @default.
- W4226227954 hasConceptScore W4226227954C55493867 @default.
- W4226227954 hasConceptScore W4226227954C63479239 @default.
- W4226227954 hasConceptScore W4226227954C76155785 @default.
- W4226227954 hasConceptScore W4226227954C97541855 @default.
- W4226227954 hasLocation W42262279541 @default.
- W4226227954 hasOpenAccess W4226227954 @default.
- W4226227954 hasPrimaryLocation W42262279541 @default.
- W4226227954 hasRelatedWork W10379689 @default.
- W4226227954 hasRelatedWork W12492200 @default.
- W4226227954 hasRelatedWork W13926559 @default.
- W4226227954 hasRelatedWork W15840176 @default.
- W4226227954 hasRelatedWork W2188409 @default.
- W4226227954 hasRelatedWork W2437517 @default.
- W4226227954 hasRelatedWork W2541924 @default.
- W4226227954 hasRelatedWork W6915741 @default.
- W4226227954 hasRelatedWork W8072425 @default.
- W4226227954 hasRelatedWork W8539471 @default.
- W4226227954 isParatext "false" @default.
- W4226227954 isRetracted "false" @default.
- W4226227954 workType "article" @default.