FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2969788501> ?p ?o ?g. }

• W2969788501 endingPage "626" @default.
• W2969788501 startingPage "610" @default.
• W2969788501 abstract "Machine learning has begun to play a central role in many applications. A multitude of these applications typically also involve datasets that are distributed across multiple computing devices/machines due to either design constraints (e.g., multi-agent and Internet-of-Things systems) or computational/privacy reasons (e.g., large-scale machine learning on smartphone data). Such applications often require the learning tasks to be carried out in a decentralized fashion, in which there is no central server that is directly connected to all nodes. In real-world decentralized settings, nodes are prone to undetected failures due to malfunctioning equipment, cyberattacks, etc., which are likely to crash non-robust learning algorithms. The focus of this paper is on robustification of decentralized learning in the presence of nodes that have undergone Byzantine failures. The Byzantine failure model allows faulty nodes to arbitrarily deviate from their intended behaviors, thereby ensuring designs of the most robust of algorithms. But the study of Byzantine resilience within decentralized learning, in contrast to distributed learning, is still in its infancy. In particular, existing Byzantine-resilient decentralized learning methods either do not scale well to large-scale machine learning models, or they lack statistical convergence guarantees that help characterize their generalization errors. In this paper, a scalable, Byzantine-resilient decentralized machine learning framework termed <b>B</b>yzantine-<b>r</b>es<b>i</b>lient <b>d</b>ecentralized <b>g</b>radient d<b>e</b>scent (BRIDGE) is introduced. Algorithmic and statistical convergence guarantees for one variant of BRIDGE are also provided in the paper for both strongly convex problems and a class of nonconvex problems. In addition, large-scale decentralized learning experiments are used to establish that the BRIDGE framework is scalable and it delivers competitive results for Byzantine-resilient convex and nonconvex learning." @default.
• W2969788501 created "2019-08-29" @default.
• W2969788501 creator A5002754104 @default.
• W2969788501 creator A5008353732 @default.
• W2969788501 creator A5028718006 @default.
• W2969788501 date "2022-01-01" @default.
• W2969788501 modified "2023-10-15" @default.
• W2969788501 title "BRIDGE: Byzantine-Resilient Decentralized Gradient Descent" @default.
• W2969788501 cites W1482321356 @default.
• W2969788501 cites W1520916864 @default.
• W2969788501 cites W1520991158 @default.
• W2969788501 cites W1873332500 @default.
• W2969788501 cites W1944672 @default.
• W2969788501 cites W1965907328 @default.
• W2969788501 cites W1970063032 @default.
• W2969788501 cites W1973184571 @default.
• W2969788501 cites W1982545958 @default.
• W2969788501 cites W2015513948 @default.
• W2969788501 cites W2020915989 @default.
• W2969788501 cites W2025375132 @default.
• W2969788501 cites W2035362408 @default.
• W2969788501 cites W2044212084 @default.
• W2969788501 cites W2056903984 @default.
• W2969788501 cites W2066332749 @default.
• W2969788501 cites W2072128103 @default.
• W2969788501 cites W2087966294 @default.
• W2969788501 cites W2100953052 @default.
• W2969788501 cites W2103012681 @default.
• W2969788501 cites W2107568926 @default.
• W2969788501 cites W2112796928 @default.
• W2969788501 cites W2118020653 @default.
• W2969788501 cites W2123705108 @default.
• W2969788501 cites W2132291180 @default.
• W2969788501 cites W2137435346 @default.
• W2969788501 cites W2156909104 @default.
• W2969788501 cites W2164278908 @default.
• W2969788501 cites W2240332484 @default.
• W2969788501 cites W2296319761 @default.
• W2969788501 cites W2490498838 @default.
• W2969788501 cites W2537046506 @default.
• W2969788501 cites W2614254310 @default.
• W2969788501 cites W2752689052 @default.
• W2969788501 cites W2798551148 @default.
• W2969788501 cites W2804948787 @default.
• W2969788501 cites W2887011391 @default.
• W2969788501 cites W2896689993 @default.
• W2969788501 cites W2940527161 @default.
• W2969788501 cites W2963165390 @default.
• W2969788501 cites W2964121744 @default.
• W2969788501 cites W2964183442 @default.
• W2969788501 cites W2964261056 @default.
• W2969788501 cites W2964275427 @default.
• W2969788501 cites W2964279249 @default.
• W2969788501 cites W3101352448 @default.
• W2969788501 cites W3101665129 @default.
• W2969788501 cites W3137092842 @default.
• W2969788501 cites W2402218985 @default.
• W2969788501 doi "https://doi.org/10.1109/tsipn.2022.3188456" @default.
• W2969788501 hasPublicationYear "2022" @default.
• W2969788501 type Work @default.
• W2969788501 sameAs 2969788501 @default.
• W2969788501 citedByCount "26" @default.
• W2969788501 countsByYear W29697885012019 @default.
• W2969788501 countsByYear W29697885012020 @default.
• W2969788501 countsByYear W29697885012021 @default.
• W2969788501 countsByYear W29697885012022 @default.
• W2969788501 countsByYear W29697885012023 @default.
• W2969788501 crossrefType "journal-article" @default.
• W2969788501 hasAuthorship W2969788501A5002754104 @default.
• W2969788501 hasAuthorship W2969788501A5008353732 @default.
• W2969788501 hasAuthorship W2969788501A5028718006 @default.
• W2969788501 hasBestOaLocation W29697885012 @default.
• W2969788501 hasConcept C119857082 @default.
• W2969788501 hasConcept C120314980 @default.
• W2969788501 hasConcept C154945302 @default.
• W2969788501 hasConcept C168021876 @default.
• W2969788501 hasConcept C17532199 @default.
• W2969788501 hasConcept C41008148 @default.
• W2969788501 hasConcept C48044578 @default.
• W2969788501 hasConcept C63540848 @default.
• W2969788501 hasConcept C77088390 @default.
• W2969788501 hasConceptScore W2969788501C119857082 @default.
• W2969788501 hasConceptScore W2969788501C120314980 @default.
• W2969788501 hasConceptScore W2969788501C154945302 @default.
• W2969788501 hasConceptScore W2969788501C168021876 @default.
• W2969788501 hasConceptScore W2969788501C17532199 @default.
• W2969788501 hasConceptScore W2969788501C41008148 @default.
• W2969788501 hasConceptScore W2969788501C48044578 @default.
• W2969788501 hasConceptScore W2969788501C63540848 @default.
• W2969788501 hasConceptScore W2969788501C77088390 @default.
• W2969788501 hasFunder F4320335353 @default.
• W2969788501 hasFunder F4320338281 @default.
• W2969788501 hasLocation W29697885011 @default.
• W2969788501 hasLocation W29697885012 @default.
• W2969788501 hasOpenAccess W2969788501 @default.
• W2969788501 hasPrimaryLocation W29697885011 @default.
• W2969788501 hasRelatedWork W1698117324 @default.
• W2969788501 hasRelatedWork W2109232026 @default.

• next