FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3110483780> ?p ?o ?g. }

• W3110483780 abstract "The label noise transition matrix $T$, reflecting the probabilities that true labels flip into noisy ones, is of vital importance to model label noise and design statistically consistent classifiers. The traditional transition matrix is limited to model closed-set label noise, where noisy training data has true class labels within the noisy label set. It is unfitted to employ such a transition matrix to model open-set label noise, where some true class labels are outside the noisy label set. Thus when considering a more realistic situation, i.e., both closed-set and open-set label noise occurs, existing methods will undesirably give biased solutions. Besides, the traditional transition matrix is limited to model instance-independent label noise, which may not perform well in practice. In this paper, we focus on learning under the mixed closed-set and open-set label noise. We address the aforementioned issues by extending the traditional transition matrix to be able to model mixed label noise, and further to the cluster-dependent transition matrix to better approximate the instance-dependent label noise in real-world applications. We term the proposed transition matrix as the cluster-dependent extended transition matrix. An unbiased estimator (i.e., extended $T$-estimator) has been designed to estimate the cluster-dependent extended transition matrix by only exploiting the noisy data. Comprehensive synthetic and real experiments validate that our method can better model the mixed label noise, following its more robust performance than the prior state-of-the-art label-noise learning methods." @default.
• W3110483780 created "2020-12-07" @default.
• W3110483780 creator A5034219615 @default.
• W3110483780 creator A5035788071 @default.
• W3110483780 creator A5047240103 @default.
• W3110483780 creator A5065250332 @default.
• W3110483780 creator A5068087395 @default.
• W3110483780 creator A5073110676 @default.
• W3110483780 creator A5086045569 @default.
• W3110483780 date "2020-12-02" @default.
• W3110483780 modified "2023-09-27" @default.
• W3110483780 title "Extended T: Learning with Mixed Closed-set and Open-set Noisy Labels." @default.
• W3110483780 cites W1514928307 @default.
• W3110483780 cites W1782590233 @default.
• W3110483780 cites W1964820473 @default.
• W3110483780 cites W2011430131 @default.
• W3110483780 cites W2113290770 @default.
• W3110483780 cites W2147614625 @default.
• W3110483780 cites W2156387975 @default.
• W3110483780 cites W2165874743 @default.
• W3110483780 cites W2302255633 @default.
• W3110483780 cites W2335728318 @default.
• W3110483780 cites W2515770085 @default.
• W3110483780 cites W2577784528 @default.
• W3110483780 cites W2603986758 @default.
• W3110483780 cites W2663800299 @default.
• W3110483780 cites W2741430497 @default.
• W3110483780 cites W2752828042 @default.
• W3110483780 cites W2887500939 @default.
• W3110483780 cites W2887842788 @default.
• W3110483780 cites W2912829604 @default.
• W3110483780 cites W2953791858 @default.
• W3110483780 cites W2962852342 @default.
• W3110483780 cites W2963081269 @default.
• W3110483780 cites W2963371670 @default.
• W3110483780 cites W2963644303 @default.
• W3110483780 cites W2963735582 @default.
• W3110483780 cites W2963759070 @default.
• W3110483780 cites W2963839617 @default.
• W3110483780 cites W2963909971 @default.
• W3110483780 cites W2964096266 @default.
• W3110483780 cites W2964292098 @default.
• W3110483780 cites W2964309657 @default.
• W3110483780 cites W2969985801 @default.
• W3110483780 cites W2970038028 @default.
• W3110483780 cites W2970711760 @default.
• W3110483780 cites W2971193222 @default.
• W3110483780 cites W2981873476 @default.
• W3110483780 cites W2981952612 @default.
• W3110483780 cites W2985817549 @default.
• W3110483780 cites W2995315671 @default.
• W3110483780 cites W2995624272 @default.
• W3110483780 cites W2996108195 @default.
• W3110483780 cites W2998375045 @default.
• W3110483780 cites W3005731330 @default.
• W3110483780 cites W3009419078 @default.
• W3110483780 cites W3010507055 @default.
• W3110483780 cites W303151045 @default.
• W3110483780 cites W3034266240 @default.
• W3110483780 cites W3034552680 @default.
• W3110483780 cites W3034950264 @default.
• W3110483780 cites W3034954777 @default.
• W3110483780 cites W3035314656 @default.
• W3110483780 cites W3035429562 @default.
• W3110483780 cites W3035702670 @default.
• W3110483780 cites W3035753488 @default.
• W3110483780 cites W3091808362 @default.
• W3110483780 cites W3092841986 @default.
• W3110483780 cites W3098082171 @default.
• W3110483780 cites W3102433174 @default.
• W3110483780 cites W3102583815 @default.
• W3110483780 cites W3102643535 @default.
• W3110483780 cites W3103846556 @default.
• W3110483780 cites W3104255659 @default.
• W3110483780 cites W3105011493 @default.
• W3110483780 cites W3109225549 @default.
• W3110483780 cites W3118608800 @default.
• W3110483780 cites W3134631405 @default.
• W3110483780 cites W3137695714 @default.
• W3110483780 cites W3167946025 @default.
• W3110483780 hasPublicationYear "2020" @default.
• W3110483780 type Work @default.
• W3110483780 sameAs 3110483780 @default.
• W3110483780 citedByCount "7" @default.
• W3110483780 countsByYear W31104837802020 @default.
• W3110483780 countsByYear W31104837802021 @default.
• W3110483780 countsByYear W31104837802022 @default.
• W3110483780 crossrefType "posted-content" @default.
• W3110483780 hasAuthorship W3110483780A5034219615 @default.
• W3110483780 hasAuthorship W3110483780A5035788071 @default.
• W3110483780 hasAuthorship W3110483780A5047240103 @default.
• W3110483780 hasAuthorship W3110483780A5065250332 @default.
• W3110483780 hasAuthorship W3110483780A5068087395 @default.
• W3110483780 hasAuthorship W3110483780A5073110676 @default.
• W3110483780 hasAuthorship W3110483780A5086045569 @default.
• W3110483780 hasConcept C104317684 @default.
• W3110483780 hasConcept C105795698 @default.
• W3110483780 hasConcept C106487976 @default.
• W3110483780 hasConcept C11413529 @default.
• W3110483780 hasConcept C115961682 @default.

• next