Matches in SemOpenAlex for { <https://semopenalex.org/work/W3080580959> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W3080580959 abstract "The past years have witnessed the rapid development of deep reinforcement learning (DRL), which is a combination of deep learning and reinforcement learning (RL). However, the adoption of deep neural networks makes the decision-making process of DRL opaque and lacking transparency. Motivated by this, various interpretation methods for DRL have been proposed. However, those interpretation methods make an implicit assumption that they are performed in a reliable and secure environment. In practice, sequential agent-environment interactions expose the DRL algorithms and their corresponding downstream interpretations to extra adversarial risk. In spite of the prevalence of malicious attacks, there is no existing work studying the possibility and feasibility of malicious attacks against DRL interpretations. To bridge this gap, in this paper, we investigate the vulnerability of DRL interpretation methods. Specifically, we introduce the first study of the adversarial attacks against DRL interpretations, and propose an optimization framework based on which the optimal adversarial attack strategy can be derived. In addition, we study the vulnerability of DRL interpretation methods to the model poisoning attacks, and present an algorithmic framework to rigorously formulate the proposed model poisoning attack. Finally, we conduct both theoretical analysis and extensive experiments to validate the effectiveness of the proposed malicious attacks against DRL interpretations." @default.
- W3080580959 created "2020-09-01" @default.
- W3080580959 creator A5013588572 @default.
- W3080580959 creator A5016035883 @default.
- W3080580959 creator A5021218766 @default.
- W3080580959 creator A5032616889 @default.
- W3080580959 creator A5067853952 @default.
- W3080580959 date "2020-08-20" @default.
- W3080580959 modified "2023-10-16" @default.
- W3080580959 title "Malicious Attacks against Deep Reinforcement Learning Interpretations" @default.
- W3080580959 cites W2788102893 @default.
- W3080580959 cites W2809622771 @default.
- W3080580959 cites W2891830784 @default.
- W3080580959 cites W2962755762 @default.
- W3080580959 cites W2962843949 @default.
- W3080580959 doi "https://doi.org/10.1145/3394486.3403089" @default.
- W3080580959 hasPublicationYear "2020" @default.
- W3080580959 type Work @default.
- W3080580959 sameAs 3080580959 @default.
- W3080580959 citedByCount "13" @default.
- W3080580959 countsByYear W30805809592020 @default.
- W3080580959 countsByYear W30805809592021 @default.
- W3080580959 countsByYear W30805809592022 @default.
- W3080580959 countsByYear W30805809592023 @default.
- W3080580959 crossrefType "proceedings-article" @default.
- W3080580959 hasAuthorship W3080580959A5013588572 @default.
- W3080580959 hasAuthorship W3080580959A5016035883 @default.
- W3080580959 hasAuthorship W3080580959A5021218766 @default.
- W3080580959 hasAuthorship W3080580959A5032616889 @default.
- W3080580959 hasAuthorship W3080580959A5067853952 @default.
- W3080580959 hasConcept C108583219 @default.
- W3080580959 hasConcept C119857082 @default.
- W3080580959 hasConcept C154945302 @default.
- W3080580959 hasConcept C199360897 @default.
- W3080580959 hasConcept C2778403875 @default.
- W3080580959 hasConcept C2780233690 @default.
- W3080580959 hasConcept C37736160 @default.
- W3080580959 hasConcept C38652104 @default.
- W3080580959 hasConcept C41008148 @default.
- W3080580959 hasConcept C527412718 @default.
- W3080580959 hasConcept C95713431 @default.
- W3080580959 hasConcept C97541855 @default.
- W3080580959 hasConceptScore W3080580959C108583219 @default.
- W3080580959 hasConceptScore W3080580959C119857082 @default.
- W3080580959 hasConceptScore W3080580959C154945302 @default.
- W3080580959 hasConceptScore W3080580959C199360897 @default.
- W3080580959 hasConceptScore W3080580959C2778403875 @default.
- W3080580959 hasConceptScore W3080580959C2780233690 @default.
- W3080580959 hasConceptScore W3080580959C37736160 @default.
- W3080580959 hasConceptScore W3080580959C38652104 @default.
- W3080580959 hasConceptScore W3080580959C41008148 @default.
- W3080580959 hasConceptScore W3080580959C527412718 @default.
- W3080580959 hasConceptScore W3080580959C95713431 @default.
- W3080580959 hasConceptScore W3080580959C97541855 @default.
- W3080580959 hasLocation W30805809591 @default.
- W3080580959 hasOpenAccess W3080580959 @default.
- W3080580959 hasPrimaryLocation W30805809591 @default.
- W3080580959 hasRelatedWork W2897573479 @default.
- W3080580959 hasRelatedWork W2963115223 @default.
- W3080580959 hasRelatedWork W3046843850 @default.
- W3080580959 hasRelatedWork W4223943233 @default.
- W3080580959 hasRelatedWork W4225586443 @default.
- W3080580959 hasRelatedWork W4367364209 @default.
- W3080580959 hasRelatedWork W4379255972 @default.
- W3080580959 hasRelatedWork W4380075502 @default.
- W3080580959 hasRelatedWork W4383468834 @default.
- W3080580959 hasRelatedWork W4383955378 @default.
- W3080580959 isParatext "false" @default.
- W3080580959 isRetracted "false" @default.
- W3080580959 magId "3080580959" @default.
- W3080580959 workType "article" @default.