Matches in SemOpenAlex for { <https://semopenalex.org/work/W2942718970> ?p ?o ?g. }
- W2942718970 abstract "Learning disentangled representations is considered a cornerstone problem in representation learning. Recently, Locatello et al. (2019) demonstrated that unsupervised disentanglement learning without inductive biases is theoretically impossible and that existing inductive biases and unsupervised methods do not allow to consistently learn disentangled representations. However, in many practical settings, one might have access to a limited amount of supervision, for example through manual labeling of (some) factors of variation in a few training examples. In this paper, we investigate the impact of such supervision on state-of-the-art disentanglement methods and perform a large scale study, training over 52000 models under well-defined and reproducible experimental conditions. We observe that a small number of labeled examples (0.01--0.5% of the data set), with potentially imprecise and incomplete labels, is sufficient to perform model selection on state-of-the-art unsupervised models. Further, we investigate the benefit of incorporating supervision into the training process. Overall, we empirically validate that with little and imprecise supervision it is possible to reliably learn disentangled representations." @default.
- W2942718970 created "2019-05-09" @default.
- W2942718970 creator A5018488407 @default.
- W2942718970 creator A5035416263 @default.
- W2942718970 creator A5044005697 @default.
- W2942718970 creator A5073157306 @default.
- W2942718970 creator A5088082340 @default.
- W2942718970 creator A5089890773 @default.
- W2942718970 date "2019-05-03" @default.
- W2942718970 modified "2023-09-27" @default.
- W2942718970 title "Disentangling Factors of Variation Using Few Labels" @default.
- W2942718970 cites W1524326598 @default.
- W2942718970 cites W1574202531 @default.
- W2942718970 cites W1691728462 @default.
- W2942718970 cites W1782879880 @default.
- W2942718970 cites W1912570122 @default.
- W2942718970 cites W1915328033 @default.
- W2942718970 cites W1959608418 @default.
- W2942718970 cites W1994618660 @default.
- W2942718970 cites W2012762214 @default.
- W2942718970 cites W2099741732 @default.
- W2942718970 cites W2105728138 @default.
- W2942718970 cites W2108501770 @default.
- W2942718970 cites W2124101779 @default.
- W2942718970 cites W2134352661 @default.
- W2942718970 cites W2134557905 @default.
- W2942718970 cites W2144020560 @default.
- W2942718970 cites W2157617585 @default.
- W2942718970 cites W2163922914 @default.
- W2942718970 cites W2184218725 @default.
- W2942718970 cites W2188956040 @default.
- W2942718970 cites W2261396597 @default.
- W2942718970 cites W2281112906 @default.
- W2942718970 cites W2559823555 @default.
- W2942718970 cites W2613634265 @default.
- W2942718970 cites W2739083961 @default.
- W2942718970 cites W2739693027 @default.
- W2942718970 cites W2753738274 @default.
- W2942718970 cites W2785961484 @default.
- W2942718970 cites W2789062658 @default.
- W2942718970 cites W2796704765 @default.
- W2942718970 cites W2803390034 @default.
- W2942718970 cites W2807255204 @default.
- W2942718970 cites W2808697642 @default.
- W2942718970 cites W2823112946 @default.
- W2942718970 cites W2892210823 @default.
- W2942718970 cites W2897959392 @default.
- W2942718970 cites W2898909293 @default.
- W2942718970 cites W2902476877 @default.
- W2942718970 cites W2903538854 @default.
- W2942718970 cites W2903991321 @default.
- W2942718970 cites W2904849495 @default.
- W2942718970 cites W2905204718 @default.
- W2942718970 cites W2919115771 @default.
- W2942718970 cites W2920527457 @default.
- W2942718970 cites W2921887948 @default.
- W2942718970 cites W2948215242 @default.
- W2942718970 cites W2950362879 @default.
- W2942718970 cites W2951736998 @default.
- W2942718970 cites W2952161038 @default.
- W2942718970 cites W2962818303 @default.
- W2942718970 cites W2963045453 @default.
- W2942718970 cites W2963104724 @default.
- W2942718970 cites W2963117534 @default.
- W2942718970 cites W2963166838 @default.
- W2942718970 cites W2963201535 @default.
- W2942718970 cites W2963226019 @default.
- W2942718970 cites W2963264829 @default.
- W2942718970 cites W2963305465 @default.
- W2942718970 cites W2963366547 @default.
- W2942718970 cites W2963445340 @default.
- W2942718970 cites W2963618559 @default.
- W2942718970 cites W2963657860 @default.
- W2942718970 cites W2963762683 @default.
- W2942718970 cites W2964127395 @default.
- W2942718970 cites W2964193753 @default.
- W2942718970 cites W2965354194 @default.
- W2942718970 cites W2966661 @default.
- W2942718970 cites W2971127577 @default.
- W2942718970 cites W2989701728 @default.
- W2942718970 cites W2990376402 @default.
- W2942718970 hasPublicationYear "2019" @default.
- W2942718970 type Work @default.
- W2942718970 sameAs 2942718970 @default.
- W2942718970 citedByCount "22" @default.
- W2942718970 countsByYear W29427189702018 @default.
- W2942718970 countsByYear W29427189702019 @default.
- W2942718970 countsByYear W29427189702020 @default.
- W2942718970 countsByYear W29427189702021 @default.
- W2942718970 countsByYear W29427189702022 @default.
- W2942718970 crossrefType "posted-content" @default.
- W2942718970 hasAuthorship W2942718970A5018488407 @default.
- W2942718970 hasAuthorship W2942718970A5035416263 @default.
- W2942718970 hasAuthorship W2942718970A5044005697 @default.
- W2942718970 hasAuthorship W2942718970A5073157306 @default.
- W2942718970 hasAuthorship W2942718970A5088082340 @default.
- W2942718970 hasAuthorship W2942718970A5089890773 @default.
- W2942718970 hasConcept C111919701 @default.
- W2942718970 hasConcept C119857082 @default.
- W2942718970 hasConcept C121332964 @default.