FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2803642127> ?p ?o ?g. }

• W2803642127 endingPage "5698" @default.
• W2803642127 startingPage "5666" @default.
• W2803642127 abstract "We study binary classification in the presence of class-conditional random noise, where the learner gets to see labels that are flipped independently with some probability, and where the flip probability depends on the class. Our goal is to devise learning algorithms that are efficient and statistically consistent with respect to commonly used utility measures. In particular, we look at a family of measures motivated by their application in domains where cost-sensitive learning is necessary (for example, when there is class imbalance). In contrast to most of the existing literature on consistent classification that are limited to the classical 0-1 loss, our analysis includes more general utility measures such as the AM measure (arithmetic mean of True Positive Rate and True Negative Rate). For this problem of cost-sensitive learning under class-conditional random noise, we develop two approaches that are based on suitably modifying surrogate losses. First, we provide a simple unbiased estimator of any loss, and obtain performance bounds for empirical utility maximization in the presence of i.i.d. data with noisy labels. If the loss function satis_es a simple symmetry condition, we show that using unbiased estimator leads to an efficient algorithm for empirical maximization. Second, by leveraging a reduction of risk minimization under noisy labels to classification with weighted 0-1 loss, we suggest the use of a simple weighted surrogate loss, for which we are able to obtain strong utility bounds. This approach implies that methods already used in practice, such as biased SVM and weighted logistic regression, are provably noise-tolerant. For two practically important measures in our family, we show that the proposed methods are competitive with respect to recently proposed methods for dealing with label noise in several benchmark data sets." @default.
• W2803642127 created "2018-06-01" @default.
• W2803642127 creator A5017175274 @default.
• W2803642127 creator A5051918150 @default.
• W2803642127 creator A5053209283 @default.
• W2803642127 creator A5063459703 @default.
• W2803642127 date "2017-01-01" @default.
• W2803642127 modified "2023-09-27" @default.
• W2803642127 title "Cost-sensitive learning with noisy labels" @default.
• W2803642127 cites W1496805632 @default.
• W2803642127 cites W1580256954 @default.
• W2803642127 cites W167016754 @default.
• W2803642127 cites W1790582767 @default.
• W2803642127 cites W1873158149 @default.
• W2803642127 cites W1966771059 @default.
• W2803642127 cites W1969623397 @default.
• W2803642127 cites W1975128126 @default.
• W2803642127 cites W1977696979 @default.
• W2803642127 cites W1982032418 @default.
• W2803642127 cites W1989888201 @default.
• W2803642127 cites W1992208280 @default.
• W2803642127 cites W1994550352 @default.
• W2803642127 cites W2003212828 @default.
• W2803642127 cites W2041268952 @default.
• W2803642127 cites W2048679005 @default.
• W2803642127 cites W2050871273 @default.
• W2803642127 cites W2074950806 @default.
• W2803642127 cites W2102348129 @default.
• W2803642127 cites W2107189314 @default.
• W2803642127 cites W2109689843 @default.
• W2803642127 cites W2120497086 @default.
• W2803642127 cites W2122537498 @default.
• W2803642127 cites W2122770142 @default.
• W2803642127 cites W2123958887 @default.
• W2803642127 cites W2134510195 @default.
• W2803642127 cites W2135606414 @default.
• W2803642127 cites W2148825261 @default.
• W2803642127 cites W2151380595 @default.
• W2803642127 cites W2156622608 @default.
• W2803642127 cites W2160218441 @default.
• W2803642127 cites W2171574551 @default.
• W2803642127 cites W2187013920 @default.
• W2803642127 cites W2490901831 @default.
• W2803642127 cites W2536321907 @default.
• W2803642127 cites W2647530754 @default.
• W2803642127 cites W2804662085 @default.
• W2803642127 cites W303151045 @default.
• W2803642127 doi "https://doi.org/10.5555/3122009.3242012" @default.
• W2803642127 hasPublicationYear "2017" @default.
• W2803642127 type Work @default.
• W2803642127 sameAs 2803642127 @default.
• W2803642127 citedByCount "17" @default.
• W2803642127 countsByYear W28036421272018 @default.
• W2803642127 countsByYear W28036421272019 @default.
• W2803642127 countsByYear W28036421272020 @default.
• W2803642127 countsByYear W28036421272021 @default.
• W2803642127 countsByYear W28036421272022 @default.
• W2803642127 crossrefType "journal-article" @default.
• W2803642127 hasAuthorship W2803642127A5017175274 @default.
• W2803642127 hasAuthorship W2803642127A5051918150 @default.
• W2803642127 hasAuthorship W2803642127A5053209283 @default.
• W2803642127 hasAuthorship W2803642127A5063459703 @default.
• W2803642127 hasConcept C105795698 @default.
• W2803642127 hasConcept C107321475 @default.
• W2803642127 hasConcept C111472728 @default.
• W2803642127 hasConcept C11413529 @default.
• W2803642127 hasConcept C115961682 @default.
• W2803642127 hasConcept C119857082 @default.
• W2803642127 hasConcept C12267149 @default.
• W2803642127 hasConcept C124101348 @default.
• W2803642127 hasConcept C126255220 @default.
• W2803642127 hasConcept C138885662 @default.
• W2803642127 hasConcept C147764199 @default.
• W2803642127 hasConcept C154945302 @default.
• W2803642127 hasConcept C185429906 @default.
• W2803642127 hasConcept C2776330181 @default.
• W2803642127 hasConcept C2776502983 @default.
• W2803642127 hasConcept C2777212361 @default.
• W2803642127 hasConcept C2780009758 @default.
• W2803642127 hasConcept C2780586882 @default.
• W2803642127 hasConcept C33923547 @default.
• W2803642127 hasConcept C39891107 @default.
• W2803642127 hasConcept C41008148 @default.
• W2803642127 hasConcept C66905080 @default.
• W2803642127 hasConcept C99498987 @default.
• W2803642127 hasConceptScore W2803642127C105795698 @default.
• W2803642127 hasConceptScore W2803642127C107321475 @default.
• W2803642127 hasConceptScore W2803642127C111472728 @default.
• W2803642127 hasConceptScore W2803642127C11413529 @default.
• W2803642127 hasConceptScore W2803642127C115961682 @default.
• W2803642127 hasConceptScore W2803642127C119857082 @default.
• W2803642127 hasConceptScore W2803642127C12267149 @default.
• W2803642127 hasConceptScore W2803642127C124101348 @default.
• W2803642127 hasConceptScore W2803642127C126255220 @default.
• W2803642127 hasConceptScore W2803642127C138885662 @default.
• W2803642127 hasConceptScore W2803642127C147764199 @default.
• W2803642127 hasConceptScore W2803642127C154945302 @default.
• W2803642127 hasConceptScore W2803642127C185429906 @default.

• next