FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2945634539> ?p ?o ?g. }

• W2945634539 abstract "Deep neural networks (DNNs) have great expressive power, which can even memorize samples with wrong labels. It is vitally important to reiterate robustness and generalization in DNNs against label corruption. To this end, this paper studies the 0-1 loss, which has a monotonic relationship with an empirical adversary (reweighted) risk~citep{hu2016does}. Although the 0-1 loss has some robust properties, it is difficult to optimize. To efficiently optimize the 0-1 loss while keeping its robust properties, we propose a very simple and efficient loss, i.e. curriculum loss (CL). Our CL is a tighter upper bound of the 0-1 loss compared with conventional summation based surrogate losses. Moreover, CL can adaptively select samples for model training. As a result, our loss can be deemed as a novel perspective of curriculum sample selection strategy, which bridges a connection between curriculum learning and robust learning. Experimental results on benchmark datasets validate the robustness of the proposed loss." @default.
• W2945634539 created "2019-05-29" @default.
• W2945634539 creator A5021751767 @default.
• W2945634539 creator A5053437240 @default.
• W2945634539 date "2019-05-24" @default.
• W2945634539 modified "2023-09-27" @default.
• W2945634539 title "Curriculum Loss: Robust Learning and Generalization against Label Corruption" @default.
• W2945634539 cites W1866072925 @default.
• W2945634539 cites W1982032418 @default.
• W2945634539 cites W2089394015 @default.
• W2945634539 cites W2100913937 @default.
• W2945634539 cites W2132984949 @default.
• W2945634539 cites W2136688338 @default.
• W2945634539 cites W2256388387 @default.
• W2945634539 cites W2296073425 @default.
• W2945634539 cites W2394658926 @default.
• W2945634539 cites W2605991684 @default.
• W2945634539 cites W2665436006 @default.
• W2945634539 cites W2752971446 @default.
• W2945634539 cites W2772357980 @default.
• W2945634539 cites W2788557041 @default.
• W2945634539 cites W2914448714 @default.
• W2945634539 cites W2920182456 @default.
• W2945634539 cites W2963081269 @default.
• W2945634539 cites W2963096987 @default.
• W2945634539 cites W2963160702 @default.
• W2945634539 cites W2963644303 @default.
• W2945634539 cites W2963735582 @default.
• W2945634539 cites W2963759070 @default.
• W2945634539 cites W2963789034 @default.
• W2945634539 cites W2964274690 @default.
• W2945634539 cites W2964292098 @default.
• W2945634539 cites W3137695714 @default.
• W2945634539 cites W369786348 @default.
• W2945634539 cites W573894934 @default.
• W2945634539 cites W2962915384 @default.
• W2945634539 hasPublicationYear "2019" @default.
• W2945634539 type Work @default.
• W2945634539 sameAs 2945634539 @default.
• W2945634539 citedByCount "12" @default.
• W2945634539 countsByYear W29456345392020 @default.
• W2945634539 countsByYear W29456345392021 @default.
• W2945634539 crossrefType "posted-content" @default.
• W2945634539 hasAuthorship W2945634539A5021751767 @default.
• W2945634539 hasAuthorship W2945634539A5053437240 @default.
• W2945634539 hasConcept C104317684 @default.
• W2945634539 hasConcept C119857082 @default.
• W2945634539 hasConcept C13280743 @default.
• W2945634539 hasConcept C134306372 @default.
• W2945634539 hasConcept C145420912 @default.
• W2945634539 hasConcept C154945302 @default.
• W2945634539 hasConcept C15744967 @default.
• W2945634539 hasConcept C177148314 @default.
• W2945634539 hasConcept C185592680 @default.
• W2945634539 hasConcept C185798385 @default.
• W2945634539 hasConcept C19417346 @default.
• W2945634539 hasConcept C205649164 @default.
• W2945634539 hasConcept C30038468 @default.
• W2945634539 hasConcept C33923547 @default.
• W2945634539 hasConcept C38652104 @default.
• W2945634539 hasConcept C41008148 @default.
• W2945634539 hasConcept C41065033 @default.
• W2945634539 hasConcept C47177190 @default.
• W2945634539 hasConcept C55493867 @default.
• W2945634539 hasConcept C63479239 @default.
• W2945634539 hasConcept C72169020 @default.
• W2945634539 hasConceptScore W2945634539C104317684 @default.
• W2945634539 hasConceptScore W2945634539C119857082 @default.
• W2945634539 hasConceptScore W2945634539C13280743 @default.
• W2945634539 hasConceptScore W2945634539C134306372 @default.
• W2945634539 hasConceptScore W2945634539C145420912 @default.
• W2945634539 hasConceptScore W2945634539C154945302 @default.
• W2945634539 hasConceptScore W2945634539C15744967 @default.
• W2945634539 hasConceptScore W2945634539C177148314 @default.
• W2945634539 hasConceptScore W2945634539C185592680 @default.
• W2945634539 hasConceptScore W2945634539C185798385 @default.
• W2945634539 hasConceptScore W2945634539C19417346 @default.
• W2945634539 hasConceptScore W2945634539C205649164 @default.
• W2945634539 hasConceptScore W2945634539C30038468 @default.
• W2945634539 hasConceptScore W2945634539C33923547 @default.
• W2945634539 hasConceptScore W2945634539C38652104 @default.
• W2945634539 hasConceptScore W2945634539C41008148 @default.
• W2945634539 hasConceptScore W2945634539C41065033 @default.
• W2945634539 hasConceptScore W2945634539C47177190 @default.
• W2945634539 hasConceptScore W2945634539C55493867 @default.
• W2945634539 hasConceptScore W2945634539C63479239 @default.
• W2945634539 hasConceptScore W2945634539C72169020 @default.
• W2945634539 hasLocation W29456345391 @default.
• W2945634539 hasOpenAccess W2945634539 @default.
• W2945634539 hasPrimaryLocation W29456345391 @default.
• W2945634539 hasRelatedWork W1975128126 @default.
• W2945634539 hasRelatedWork W2102348129 @default.
• W2945634539 hasRelatedWork W2194775991 @default.
• W2945634539 hasRelatedWork W2752971446 @default.
• W2945634539 hasRelatedWork W2767029637 @default.
• W2945634539 hasRelatedWork W2903329274 @default.
• W2945634539 hasRelatedWork W2963207607 @default.
• W2945634539 hasRelatedWork W2963735582 @default.
• W2945634539 hasRelatedWork W2963759070 @default.
• W2945634539 hasRelatedWork W2964292098 @default.

• next