Matches in SemOpenAlex for { <https://semopenalex.org/work/W3200256144> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W3200256144 abstract "Federated learning (FL) has become an emerging machine learning technique lately due to its efficacy in safeguarding the client’s confidential information. Nevertheless, despite the inherent and additional privacy-preserving mechanisms (e.g., differential privacy, secure multi-party computation, etc.), the FL models are still vulnerable to various privacy-violating and security-compromising attacks (e.g., data or model poisoning) due to their numerous attack vectors which in turn, make the models either ineffective or suboptimal. Existing adversarial models focusing on untargeted model poisoning attacks are not enough stealthy and persistent at the same time because of their conflicting nature (large scale attacks are easier to detect and vice versa) and thus, remain an unsolved research problem in this adversarial learning paradigm. Considering this, in this paper, we analyze this adversarial learning process in an FL setting and show that a stealthy and persistent model poisoning attack can be conducted exploiting the differential noise. More specifically, we develop an unprecedented DP-exploited stealthy model poisoning (DeSMP) attack for FL models. Our empirical analysis on both the classification and regression tasks using two popular datasets reflects the effectiveness of the proposed DeSMP attack. Moreover, we develop a novel reinforcement learning (RL)-based defense strategy against such model poisoning attacks which can intelligently and dynamically select the privacy level of the FL models to minimize the DeSMP attack surface and facilitate the attack detection." @default.
- W3200256144 created "2021-09-27" @default.
- W3200256144 creator A5042166963 @default.
- W3200256144 creator A5044262885 @default.
- W3200256144 creator A5048107620 @default.
- W3200256144 creator A5078188257 @default.
- W3200256144 date "2021-12-01" @default.
- W3200256144 modified "2023-10-18" @default.
- W3200256144 title "DeSMP: Differential Privacy-exploited Stealthy Model Poisoning Attacks in Federated Learning" @default.
- W3200256144 cites W1873763122 @default.
- W3200256144 cites W2007339694 @default.
- W3200256144 cites W2473418344 @default.
- W3200256144 cites W2535690855 @default.
- W3200256144 cites W2912161124 @default.
- W3200256144 cites W2922269401 @default.
- W3200256144 cites W2970408908 @default.
- W3200256144 cites W2981347044 @default.
- W3200256144 cites W3008145398 @default.
- W3200256144 cites W3027176637 @default.
- W3200256144 cites W3045941597 @default.
- W3200256144 cites W3087391814 @default.
- W3200256144 cites W3100146630 @default.
- W3200256144 cites W3103802018 @default.
- W3200256144 cites W3136620885 @default.
- W3200256144 cites W3181431767 @default.
- W3200256144 cites W3200331505 @default.
- W3200256144 doi "https://doi.org/10.1109/msn53354.2021.00038" @default.
- W3200256144 hasPublicationYear "2021" @default.
- W3200256144 type Work @default.
- W3200256144 sameAs 3200256144 @default.
- W3200256144 citedByCount "9" @default.
- W3200256144 countsByYear W32002561442022 @default.
- W3200256144 countsByYear W32002561442023 @default.
- W3200256144 crossrefType "proceedings-article" @default.
- W3200256144 hasAuthorship W3200256144A5042166963 @default.
- W3200256144 hasAuthorship W3200256144A5044262885 @default.
- W3200256144 hasAuthorship W3200256144A5048107620 @default.
- W3200256144 hasAuthorship W3200256144A5078188257 @default.
- W3200256144 hasBestOaLocation W32002561442 @default.
- W3200256144 hasConcept C119857082 @default.
- W3200256144 hasConcept C124101348 @default.
- W3200256144 hasConcept C154945302 @default.
- W3200256144 hasConcept C159110408 @default.
- W3200256144 hasConcept C23130292 @default.
- W3200256144 hasConcept C2776743756 @default.
- W3200256144 hasConcept C2778403875 @default.
- W3200256144 hasConcept C2992525071 @default.
- W3200256144 hasConcept C37736160 @default.
- W3200256144 hasConcept C38652104 @default.
- W3200256144 hasConcept C41008148 @default.
- W3200256144 hasConcept C71745522 @default.
- W3200256144 hasConcept C71924100 @default.
- W3200256144 hasConcept C97541855 @default.
- W3200256144 hasConceptScore W3200256144C119857082 @default.
- W3200256144 hasConceptScore W3200256144C124101348 @default.
- W3200256144 hasConceptScore W3200256144C154945302 @default.
- W3200256144 hasConceptScore W3200256144C159110408 @default.
- W3200256144 hasConceptScore W3200256144C23130292 @default.
- W3200256144 hasConceptScore W3200256144C2776743756 @default.
- W3200256144 hasConceptScore W3200256144C2778403875 @default.
- W3200256144 hasConceptScore W3200256144C2992525071 @default.
- W3200256144 hasConceptScore W3200256144C37736160 @default.
- W3200256144 hasConceptScore W3200256144C38652104 @default.
- W3200256144 hasConceptScore W3200256144C41008148 @default.
- W3200256144 hasConceptScore W3200256144C71745522 @default.
- W3200256144 hasConceptScore W3200256144C71924100 @default.
- W3200256144 hasConceptScore W3200256144C97541855 @default.
- W3200256144 hasLocation W32002561441 @default.
- W3200256144 hasLocation W32002561442 @default.
- W3200256144 hasOpenAccess W3200256144 @default.
- W3200256144 hasPrimaryLocation W32002561441 @default.
- W3200256144 hasRelatedWork W2897573479 @default.
- W3200256144 hasRelatedWork W3013617128 @default.
- W3200256144 hasRelatedWork W3172173631 @default.
- W3200256144 hasRelatedWork W4224883155 @default.
- W3200256144 hasRelatedWork W4225586443 @default.
- W3200256144 hasRelatedWork W4225781840 @default.
- W3200256144 hasRelatedWork W4251088474 @default.
- W3200256144 hasRelatedWork W4287815799 @default.
- W3200256144 hasRelatedWork W4295806247 @default.
- W3200256144 hasRelatedWork W4385713380 @default.
- W3200256144 isParatext "false" @default.
- W3200256144 isRetracted "false" @default.
- W3200256144 magId "3200256144" @default.
- W3200256144 workType "article" @default.