FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2888934269> ?p ?o ?g. }

• W2888934269 abstract "Empirical risk minimization (ERM), with proper loss function and regularization, is the common practice of supervised classification. In this paper, we study training arbitrary (from linear to deep) binary classifier from only unlabeled (U) data by ERM. We prove that it is impossible to estimate the risk of an arbitrary binary classifier in an unbiased manner given a single set of U data, but it becomes possible given two sets of U data with different class priors. These two facts answer a fundamental question---what the minimal supervision is for training any binary classifier from only U data. Following these findings, we propose an ERM-based learning method from two sets of U data, and then prove it is consistent. Experiments demonstrate the proposed method could train deep models and outperform state-of-the-art methods for learning from two sets of U data." @default.
• W2888934269 created "2018-09-07" @default.
• W2888934269 creator A5007098537 @default.
• W2888934269 creator A5016391927 @default.
• W2888934269 creator A5049656925 @default.
• W2888934269 creator A5072744508 @default.
• W2888934269 date "2018-08-31" @default.
• W2888934269 modified "2023-10-01" @default.
• W2888934269 title "On the Minimal Supervision for Training Any Binary Classifier from Only Unlabeled Data." @default.
• W2888934269 cites W129176699 @default.
• W2888934269 cites W1480538416 @default.
• W2888934269 cites W1514928307 @default.
• W2888934269 cites W1607038179 @default.
• W2888934269 cites W1825821140 @default.
• W2888934269 cites W1836465849 @default.
• W2888934269 cites W1866935739 @default.
• W2888934269 cites W1944672 @default.
• W2888934269 cites W1982032418 @default.
• W2888934269 cites W1994616650 @default.
• W2888934269 cites W2034909454 @default.
• W2888934269 cites W2095705004 @default.
• W2888934269 cites W2096706838 @default.
• W2888934269 cites W2097482982 @default.
• W2888934269 cites W2101557761 @default.
• W2888934269 cites W2101807845 @default.
• W2888934269 cites W2104290444 @default.
• W2888934269 cites W2107189314 @default.
• W2888934269 cites W2112796928 @default.
• W2888934269 cites W2113290770 @default.
• W2888934269 cites W2113592823 @default.
• W2888934269 cites W2123958887 @default.
• W2888934269 cites W2132442585 @default.
• W2888934269 cites W2132820034 @default.
• W2888934269 cites W2137420707 @default.
• W2888934269 cites W2141416357 @default.
• W2888934269 cites W2145494108 @default.
• W2888934269 cites W2146502635 @default.
• W2888934269 cites W2148603752 @default.
• W2888934269 cites W2149982386 @default.
• W2888934269 cites W2162651021 @default.
• W2888934269 cites W2194775991 @default.
• W2888934269 cites W2335728318 @default.
• W2888934269 cites W2536321907 @default.
• W2888934269 cites W2557283755 @default.
• W2888934269 cites W2592691248 @default.
• W2888934269 cites W2750384547 @default.
• W2888934269 cites W2752971446 @default.
• W2888934269 cites W2804344481 @default.
• W2888934269 cites W2886811283 @default.
• W2888934269 cites W2950365520 @default.
• W2888934269 cites W2962762541 @default.
• W2888934269 cites W2962971234 @default.
• W2888934269 cites W2963081269 @default.
• W2888934269 cites W2963341628 @default.
• W2888934269 cites W2963371670 @default.
• W2888934269 cites W2963382180 @default.
• W2888934269 cites W2963404486 @default.
• W2888934269 cites W2963644303 @default.
• W2888934269 cites W2963735582 @default.
• W2888934269 cites W2964040467 @default.
• W2888934269 cites W2964121744 @default.
• W2888934269 cites W2964292098 @default.
• W2888934269 cites W2973562770 @default.
• W2888934269 cites W2989661724 @default.
• W2888934269 cites W3023786531 @default.
• W2888934269 cites W3118608800 @default.
• W2888934269 cites W607505555 @default.
• W2888934269 cites W2530816535 @default.
• W2888934269 hasPublicationYear "2018" @default.
• W2888934269 type Work @default.
• W2888934269 sameAs 2888934269 @default.
• W2888934269 citedByCount "0" @default.
• W2888934269 crossrefType "posted-content" @default.
• W2888934269 hasAuthorship W2888934269A5007098537 @default.
• W2888934269 hasAuthorship W2888934269A5016391927 @default.
• W2888934269 hasAuthorship W2888934269A5049656925 @default.
• W2888934269 hasAuthorship W2888934269A5072744508 @default.
• W2888934269 hasConcept C107321475 @default.
• W2888934269 hasConcept C107673813 @default.
• W2888934269 hasConcept C119857082 @default.
• W2888934269 hasConcept C12267149 @default.
• W2888934269 hasConcept C147764199 @default.
• W2888934269 hasConcept C153180895 @default.
• W2888934269 hasConcept C154945302 @default.
• W2888934269 hasConcept C177769412 @default.
• W2888934269 hasConcept C199360897 @default.
• W2888934269 hasConcept C2776135515 @default.
• W2888934269 hasConcept C2776145971 @default.
• W2888934269 hasConcept C33923547 @default.
• W2888934269 hasConcept C41008148 @default.
• W2888934269 hasConcept C48372109 @default.
• W2888934269 hasConcept C51632099 @default.
• W2888934269 hasConcept C66905080 @default.
• W2888934269 hasConcept C94375191 @default.
• W2888934269 hasConcept C95623464 @default.
• W2888934269 hasConceptScore W2888934269C107321475 @default.
• W2888934269 hasConceptScore W2888934269C107673813 @default.
• W2888934269 hasConceptScore W2888934269C119857082 @default.
• W2888934269 hasConceptScore W2888934269C12267149 @default.
• W2888934269 hasConceptScore W2888934269C147764199 @default.

• next