FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3049367159> ?p ?o ?g. }

• W3049367159 abstract "Recent studies have demonstrated that reinforcement learning (RL) agents are susceptible to adversarial manipulation, similar to vulnerabilities previously demonstrated in the supervised learning setting. While most existing work studies the problem in the context of computer vision or console games, this paper focuses on reinforcement learning in autonomous cyber defence under partial observability. We demonstrate that under the black-box setting, where the attacker has no direct access to the target RL model, causative attacks---attacks that target the training process---can poison RL agents even if the attacker only has partial observability of the environment. In addition, we propose an inversion defence method that aims to apply the opposite perturbation to that which an attacker might use to generate their adversarial samples. Our experimental results illustrate that the countermeasure can effectively reduce the impact of the causative attack, while not significantly affecting the training process in non-attack scenarios." @default.
• W3049367159 created "2020-08-21" @default.
• W3049367159 creator A5001645149 @default.
• W3049367159 creator A5017066249 @default.
• W3049367159 creator A5020908862 @default.
• W3049367159 creator A5024595046 @default.
• W3049367159 creator A5026180696 @default.
• W3049367159 creator A5060997326 @default.
• W3049367159 creator A5070030398 @default.
• W3049367159 creator A5077639290 @default.
• W3049367159 creator A5078824132 @default.
• W3049367159 date "2019-02-24" @default.
• W3049367159 modified "2023-10-18" @default.
• W3049367159 title "Adversarial Reinforcement Learning under Partial Observability in Autonomous Computer Network Defence" @default.
• W3049367159 cites W1515851193 @default.
• W3049367159 cites W1564743226 @default.
• W3049367159 cites W1673923490 @default.
• W3049367159 cites W1757796397 @default.
• W3049367159 cites W1932198206 @default.
• W3049367159 cites W1945616565 @default.
• W3049367159 cites W2007562169 @default.
• W3049367159 cites W2041008854 @default.
• W3049367159 cites W2095195675 @default.
• W3049367159 cites W2095577883 @default.
• W3049367159 cites W2112507308 @default.
• W3049367159 cites W2115211925 @default.
• W3049367159 cites W2125908420 @default.
• W3049367159 cites W2155968351 @default.
• W3049367159 cites W2174868984 @default.
• W3049367159 cites W2180612164 @default.
• W3049367159 cites W2243397390 @default.
• W3049367159 cites W2408141691 @default.
• W3049367159 cites W2410216425 @default.
• W3049367159 cites W2543927648 @default.
• W3049367159 cites W2572659264 @default.
• W3049367159 cites W2594867206 @default.
• W3049367159 cites W2598549816 @default.
• W3049367159 cites W2603766943 @default.
• W3049367159 cites W2604394466 @default.
• W3049367159 cites W2612372205 @default.
• W3049367159 cites W2613144983 @default.
• W3049367159 cites W2620008831 @default.
• W3049367159 cites W2620038827 @default.
• W3049367159 cites W2625220439 @default.
• W3049367159 cites W2626801932 @default.
• W3049367159 cites W2747237166 @default.
• W3049367159 cites W2763443044 @default.
• W3049367159 cites W2768209273 @default.
• W3049367159 cites W2773525213 @default.
• W3049367159 cites W2773691349 @default.
• W3049367159 cites W2787708942 @default.
• W3049367159 cites W2803678876 @default.
• W3049367159 cites W2803831897 @default.
• W3049367159 cites W2883567023 @default.
• W3049367159 cites W2887432790 @default.
• W3049367159 cites W2949103145 @default.
• W3049367159 cites W2950468330 @default.
• W3049367159 cites W2951065069 @default.
• W3049367159 cites W2952477728 @default.
• W3049367159 cites W2963431851 @default.
• W3049367159 cites W2963477884 @default.
• W3049367159 cites W2963857521 @default.
• W3049367159 cites W2964043796 @default.
• W3049367159 cites W2964253222 @default.
• W3049367159 cites W2967595108 @default.
• W3049367159 doi "https://doi.org/10.48550/arxiv.1902.09062" @default.
• W3049367159 hasPublicationYear "2019" @default.
• W3049367159 type Work @default.
• W3049367159 sameAs 3049367159 @default.
• W3049367159 citedByCount "3" @default.
• W3049367159 countsByYear W30493671592019 @default.
• W3049367159 countsByYear W30493671592021 @default.
• W3049367159 crossrefType "posted-content" @default.
• W3049367159 hasAuthorship W3049367159A5001645149 @default.
• W3049367159 hasAuthorship W3049367159A5017066249 @default.
• W3049367159 hasAuthorship W3049367159A5020908862 @default.
• W3049367159 hasAuthorship W3049367159A5024595046 @default.
• W3049367159 hasAuthorship W3049367159A5026180696 @default.
• W3049367159 hasAuthorship W3049367159A5060997326 @default.
• W3049367159 hasAuthorship W3049367159A5070030398 @default.
• W3049367159 hasAuthorship W3049367159A5077639290 @default.
• W3049367159 hasAuthorship W3049367159A5078824132 @default.
• W3049367159 hasBestOaLocation W30493671591 @default.
• W3049367159 hasConcept C119857082 @default.
• W3049367159 hasConcept C127413603 @default.
• W3049367159 hasConcept C146978453 @default.
• W3049367159 hasConcept C151730666 @default.
• W3049367159 hasConcept C154945302 @default.
• W3049367159 hasConcept C21593369 @default.
• W3049367159 hasConcept C2779343474 @default.
• W3049367159 hasConcept C28826006 @default.
• W3049367159 hasConcept C33923547 @default.
• W3049367159 hasConcept C36299963 @default.
• W3049367159 hasConcept C37736160 @default.
• W3049367159 hasConcept C38652104 @default.
• W3049367159 hasConcept C41008148 @default.
• W3049367159 hasConcept C86803240 @default.
• W3049367159 hasConcept C97541855 @default.
• W3049367159 hasConceptScore W3049367159C119857082 @default.
• W3049367159 hasConceptScore W3049367159C127413603 @default.

• next