Matches in SemOpenAlex for { <https://semopenalex.org/work/W3196347836> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W3196347836 abstract "The Bayes error rate (BER) is a fundamental concept in machine learning that quantifies the best possible accuracy any classifier can achieve on a fixed probability distribution. Despite years of research on building estimators of lower and upper bounds for the BER, these were usually compared only on synthetic datasets with known probability distributions, leaving two key questions unanswered: (1) How well do they perform on real-world datasets?, and (2) How practical are they? Answering these is not trivial. Apart from the obvious challenge of an unknown BER for real-world datasets, there are two main aspects any BER estimator needs to overcome in order to be applicable in real-world settings: (1) the computational and sample complexity, and (2) the sensitivity and selection of hyper-parameters. In this work, we propose FeeBee, the first principled framework for analyzing and comparing BER estimators on any modern real-world dataset with unknown probability distribution. We achieve this by injecting a controlled amount of label noise and performing multiple evaluations on a series of different noise levels, supported by a theoretical result which allows drawing conclusions about the evolution of the BER. By implementing and analyzing 7 multi-class BER estimators on 6 commonly used datasets of the computer vision and NLP domains, FeeBee allows a thorough study of these estimators, clearly identifying strengths and weaknesses of each, whilst being easily deployable on any future BER estimator." @default.
- W3196347836 created "2021-09-13" @default.
- W3196347836 creator A5020051246 @default.
- W3196347836 creator A5025749600 @default.
- W3196347836 creator A5041232755 @default.
- W3196347836 creator A5063262952 @default.
- W3196347836 date "2021-08-30" @default.
- W3196347836 modified "2023-10-01" @default.
- W3196347836 title "Evaluating Bayes Error Estimators on Read-World Datasets with FeeBee" @default.
- W3196347836 cites W1965090383 @default.
- W3196347836 cites W1967932538 @default.
- W3196347836 cites W1983963519 @default.
- W3196347836 cites W1999271084 @default.
- W3196347836 cites W2050360441 @default.
- W3196347836 cites W2106123800 @default.
- W3196347836 cites W2122111042 @default.
- W3196347836 cites W2137827358 @default.
- W3196347836 cites W2161880915 @default.
- W3196347836 cites W2162207080 @default.
- W3196347836 cites W2163055872 @default.
- W3196347836 cites W2163605009 @default.
- W3196347836 cites W2218423201 @default.
- W3196347836 cites W3024540227 @default.
- W3196347836 cites W3086566473 @default.
- W3196347836 cites W3094502228 @default.
- W3196347836 cites W3100361261 @default.
- W3196347836 cites W3100475565 @default.
- W3196347836 cites W3122542623 @default.
- W3196347836 cites W3194633457 @default.
- W3196347836 hasPublicationYear "2021" @default.
- W3196347836 type Work @default.
- W3196347836 sameAs 3196347836 @default.
- W3196347836 citedByCount "0" @default.
- W3196347836 crossrefType "posted-content" @default.
- W3196347836 hasAuthorship W3196347836A5020051246 @default.
- W3196347836 hasAuthorship W3196347836A5025749600 @default.
- W3196347836 hasAuthorship W3196347836A5041232755 @default.
- W3196347836 hasAuthorship W3196347836A5063262952 @default.
- W3196347836 hasConcept C105795698 @default.
- W3196347836 hasConcept C107673813 @default.
- W3196347836 hasConcept C119857082 @default.
- W3196347836 hasConcept C124101348 @default.
- W3196347836 hasConcept C149441793 @default.
- W3196347836 hasConcept C154945302 @default.
- W3196347836 hasConcept C185429906 @default.
- W3196347836 hasConcept C207201462 @default.
- W3196347836 hasConcept C33923547 @default.
- W3196347836 hasConcept C41008148 @default.
- W3196347836 hasConcept C95623464 @default.
- W3196347836 hasConceptScore W3196347836C105795698 @default.
- W3196347836 hasConceptScore W3196347836C107673813 @default.
- W3196347836 hasConceptScore W3196347836C119857082 @default.
- W3196347836 hasConceptScore W3196347836C124101348 @default.
- W3196347836 hasConceptScore W3196347836C149441793 @default.
- W3196347836 hasConceptScore W3196347836C154945302 @default.
- W3196347836 hasConceptScore W3196347836C185429906 @default.
- W3196347836 hasConceptScore W3196347836C207201462 @default.
- W3196347836 hasConceptScore W3196347836C33923547 @default.
- W3196347836 hasConceptScore W3196347836C41008148 @default.
- W3196347836 hasConceptScore W3196347836C95623464 @default.
- W3196347836 hasLocation W31963478361 @default.
- W3196347836 hasOpenAccess W3196347836 @default.
- W3196347836 hasPrimaryLocation W31963478361 @default.
- W3196347836 hasRelatedWork W119609261 @default.
- W3196347836 hasRelatedWork W1966705109 @default.
- W3196347836 hasRelatedWork W2093362081 @default.
- W3196347836 hasRelatedWork W2138252504 @default.
- W3196347836 hasRelatedWork W2138438505 @default.
- W3196347836 hasRelatedWork W2403597675 @default.
- W3196347836 hasRelatedWork W2493927023 @default.
- W3196347836 hasRelatedWork W2504431993 @default.
- W3196347836 hasRelatedWork W2559982938 @default.
- W3196347836 hasRelatedWork W2888328726 @default.
- W3196347836 hasRelatedWork W2955989140 @default.
- W3196347836 hasRelatedWork W2963020076 @default.
- W3196347836 hasRelatedWork W2982391123 @default.
- W3196347836 hasRelatedWork W3003830418 @default.
- W3196347836 hasRelatedWork W3098711891 @default.
- W3196347836 hasRelatedWork W3106005960 @default.
- W3196347836 hasRelatedWork W3122321081 @default.
- W3196347836 hasRelatedWork W3135386061 @default.
- W3196347836 hasRelatedWork W3206020884 @default.
- W3196347836 hasRelatedWork W940094962 @default.
- W3196347836 isParatext "false" @default.
- W3196347836 isRetracted "false" @default.
- W3196347836 magId "3196347836" @default.
- W3196347836 workType "article" @default.