FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4311721665> ?p ?o ?g. }

• W4311721665 endingPage "130656" @default.
• W4311721665 startingPage "130642" @default.
• W4311721665 abstract "Federated learning (FL) goes beyond traditional, centralized machine learning by distributing model training among a large collection of edge clients. These clients cooperatively train a global, e.g., cloud-hosted, model without disclosing their local, private training data. The global model is then shared among all the participants which use it for local predictions. This paper proves that FL systems can be turned into covert channels to implement a stealth communication infrastructure. The main intuition is that, during federated training, a malicious sender can poison the global model by submitting purposely crafted examples. Although the effect of the model poisoning is negligible to other participants and does not alter the overall model performance, it can be observed by a malicious receiver and used to transmit a sequence of bits. We mounted our attack on an FL system to verify its feasibility. Experimental evidence shows that this covert channel is reliable, efficient, and extremely hard to counter. These results highlight that our new attacker model threatens FL infrastructures." @default.
• W4311721665 created "2022-12-28" @default.
• W4311721665 creator A5011299100 @default.
• W4311721665 creator A5036237095 @default.
• W4311721665 creator A5040739994 @default.
• W4311721665 creator A5071248912 @default.
• W4311721665 date "2022-01-01" @default.
• W4311721665 modified "2023-10-01" @default.
• W4311721665 title "Turning Federated Learning Systems Into Covert Channels" @default.
• W4311721665 cites W1480376833 @default.
• W4311721665 cites W1993704367 @default.
• W4311721665 cites W2007562169 @default.
• W4311721665 cites W2051267297 @default.
• W4311721665 cites W2063100290 @default.
• W4311721665 cites W2095577883 @default.
• W4311721665 cites W2099137371 @default.
• W4311721665 cites W2112796928 @default.
• W4311721665 cites W2137200163 @default.
• W4311721665 cites W2147800946 @default.
• W4311721665 cites W2151298633 @default.
• W4311721665 cites W2293768274 @default.
• W4311721665 cites W2336566325 @default.
• W4311721665 cites W2396115941 @default.
• W4311721665 cites W2535690855 @default.
• W4311721665 cites W2603766943 @default.
• W4311721665 cites W2613314732 @default.
• W4311721665 cites W2753783305 @default.
• W4311721665 cites W2892160417 @default.
• W4311721665 cites W2897830718 @default.
• W4311721665 cites W2897865027 @default.
• W4311721665 cites W2930926105 @default.
• W4311721665 cites W2942091739 @default.
• W4311721665 cites W2962763344 @default.
• W4311721665 cites W2962835266 @default.
• W4311721665 cites W2963456518 @default.
• W4311721665 cites W2963844355 @default.
• W4311721665 cites W2963857521 @default.
• W4311721665 cites W2964151798 @default.
• W4311721665 cites W2973628901 @default.
• W4311721665 cites W2981446616 @default.
• W4311721665 cites W2982302101 @default.
• W4311721665 cites W3016560828 @default.
• W4311721665 cites W3016632787 @default.
• W4311721665 cites W3018464563 @default.
• W4311721665 cites W3021026170 @default.
• W4311721665 cites W3030742901 @default.
• W4311721665 cites W3037806987 @default.
• W4311721665 cites W3087391814 @default.
• W4311721665 cites W3111358184 @default.
• W4311721665 cites W3114953370 @default.
• W4311721665 cites W3201824817 @default.
• W4311721665 cites W4247200422 @default.
• W4311721665 cites W4294106961 @default.
• W4311721665 cites W9657784 @default.
• W4311721665 doi "https://doi.org/10.1109/access.2022.3229124" @default.
• W4311721665 hasPublicationYear "2022" @default.
• W4311721665 type Work @default.
• W4311721665 citedByCount "1" @default.
• W4311721665 countsByYear W43117216652023 @default.
• W4311721665 crossrefType "journal-article" @default.
• W4311721665 hasAuthorship W4311721665A5011299100 @default.
• W4311721665 hasAuthorship W4311721665A5036237095 @default.
• W4311721665 hasAuthorship W4311721665A5040739994 @default.
• W4311721665 hasAuthorship W4311721665A5071248912 @default.
• W4311721665 hasBestOaLocation W43117216652 @default.
• W4311721665 hasConcept C103377522 @default.
• W4311721665 hasConcept C111472728 @default.
• W4311721665 hasConcept C111919701 @default.
• W4311721665 hasConcept C127162648 @default.
• W4311721665 hasConcept C132010649 @default.
• W4311721665 hasConcept C138236772 @default.
• W4311721665 hasConcept C138885662 @default.
• W4311721665 hasConcept C154945302 @default.
• W4311721665 hasConcept C162307627 @default.
• W4311721665 hasConcept C184842701 @default.
• W4311721665 hasConcept C198104137 @default.
• W4311721665 hasConcept C2779338814 @default.
• W4311721665 hasConcept C29024540 @default.
• W4311721665 hasConcept C2992525071 @default.
• W4311721665 hasConcept C31258907 @default.
• W4311721665 hasConcept C38652104 @default.
• W4311721665 hasConcept C41008148 @default.
• W4311721665 hasConcept C41895202 @default.
• W4311721665 hasConcept C79974875 @default.
• W4311721665 hasConceptScore W4311721665C103377522 @default.
• W4311721665 hasConceptScore W4311721665C111472728 @default.
• W4311721665 hasConceptScore W4311721665C111919701 @default.
• W4311721665 hasConceptScore W4311721665C127162648 @default.
• W4311721665 hasConceptScore W4311721665C132010649 @default.
• W4311721665 hasConceptScore W4311721665C138236772 @default.
• W4311721665 hasConceptScore W4311721665C138885662 @default.
• W4311721665 hasConceptScore W4311721665C154945302 @default.
• W4311721665 hasConceptScore W4311721665C162307627 @default.
• W4311721665 hasConceptScore W4311721665C184842701 @default.
• W4311721665 hasConceptScore W4311721665C198104137 @default.
• W4311721665 hasConceptScore W4311721665C2779338814 @default.
• W4311721665 hasConceptScore W4311721665C29024540 @default.
• W4311721665 hasConceptScore W4311721665C2992525071 @default.

• next