Matches in SemOpenAlex for { <https://semopenalex.org/work/W918148966> ?p ?o ?g. }
- W918148966 endingPage "1683" @default.
- W918148966 startingPage "1675" @default.
- W918148966 abstract "Careful tuning of a regularization parameter is indispensable in many machine learning tasks because it has a significant impact on generalization performances. Nevertheless, current practice of regularization parameter tuning is more of an art than a science, e.g., it is hard to tell how many grid-points would be needed in cross-validation (CV) for obtaining a solution with sufficiently small CV error. In this paper we propose a novel framework for computing a lower bound of the CV errors as a function of the regularization parameter, which we call regularization path of CV error lower bounds. The proposed framework can be used for providing a theoretical approximation guarantee on a set of solutions in the sense that how far the CV error of the current best solution could be away from best possible CV error in the entire range of the regularization parameters. Our numerical experiments demonstrate that a theoretically guaranteed choice of a regularization parameter in the above sense is possible with reasonable computational costs." @default.
- W918148966 created "2016-06-24" @default.
- W918148966 creator A5002160178 @default.
- W918148966 creator A5003668143 @default.
- W918148966 creator A5004119000 @default.
- W918148966 creator A5081482638 @default.
- W918148966 date "2015-12-07" @default.
- W918148966 modified "2023-09-24" @default.
- W918148966 title "Regularization path of cross-validation error lower bounds" @default.
- W918148966 cites W135566594 @default.
- W918148966 cites W1536816303 @default.
- W918148966 cites W1539150806 @default.
- W918148966 cites W1543911290 @default.
- W918148966 cites W2061810505 @default.
- W918148966 cites W2063978378 @default.
- W918148966 cites W2097998348 @default.
- W918148966 cites W2099680562 @default.
- W918148966 cites W2100268245 @default.
- W918148966 cites W2101399145 @default.
- W918148966 cites W2112495313 @default.
- W918148966 cites W2113584252 @default.
- W918148966 cites W2118585731 @default.
- W918148966 cites W2129191766 @default.
- W918148966 cites W2129697675 @default.
- W918148966 cites W2131241448 @default.
- W918148966 cites W2133958955 @default.
- W918148966 cites W2137475104 @default.
- W918148966 cites W2146258363 @default.
- W918148966 cites W2147898188 @default.
- W918148966 cites W2153635508 @default.
- W918148966 cites W2156909104 @default.
- W918148966 cites W2157532876 @default.
- W918148966 cites W2158001550 @default.
- W918148966 cites W2798766386 @default.
- W918148966 cites W607505555 @default.
- W918148966 cites W740415 @default.
- W918148966 hasPublicationYear "2015" @default.
- W918148966 type Work @default.
- W918148966 sameAs 918148966 @default.
- W918148966 citedByCount "2" @default.
- W918148966 countsByYear W9181489662018 @default.
- W918148966 crossrefType "proceedings-article" @default.
- W918148966 hasAuthorship W918148966A5002160178 @default.
- W918148966 hasAuthorship W918148966A5003668143 @default.
- W918148966 hasAuthorship W918148966A5004119000 @default.
- W918148966 hasAuthorship W918148966A5081482638 @default.
- W918148966 hasConcept C11413529 @default.
- W918148966 hasConcept C117765406 @default.
- W918148966 hasConcept C119857082 @default.
- W918148966 hasConcept C126255220 @default.
- W918148966 hasConcept C134306372 @default.
- W918148966 hasConcept C135252773 @default.
- W918148966 hasConcept C141718189 @default.
- W918148966 hasConcept C152442038 @default.
- W918148966 hasConcept C154945302 @default.
- W918148966 hasConcept C2776135515 @default.
- W918148966 hasConcept C27872270 @default.
- W918148966 hasConcept C28826006 @default.
- W918148966 hasConcept C33923547 @default.
- W918148966 hasConcept C41008148 @default.
- W918148966 hasConcept C50644808 @default.
- W918148966 hasConcept C77553402 @default.
- W918148966 hasConceptScore W918148966C11413529 @default.
- W918148966 hasConceptScore W918148966C117765406 @default.
- W918148966 hasConceptScore W918148966C119857082 @default.
- W918148966 hasConceptScore W918148966C126255220 @default.
- W918148966 hasConceptScore W918148966C134306372 @default.
- W918148966 hasConceptScore W918148966C135252773 @default.
- W918148966 hasConceptScore W918148966C141718189 @default.
- W918148966 hasConceptScore W918148966C152442038 @default.
- W918148966 hasConceptScore W918148966C154945302 @default.
- W918148966 hasConceptScore W918148966C2776135515 @default.
- W918148966 hasConceptScore W918148966C27872270 @default.
- W918148966 hasConceptScore W918148966C28826006 @default.
- W918148966 hasConceptScore W918148966C33923547 @default.
- W918148966 hasConceptScore W918148966C41008148 @default.
- W918148966 hasConceptScore W918148966C50644808 @default.
- W918148966 hasConceptScore W918148966C77553402 @default.
- W918148966 hasLocation W9181489661 @default.
- W918148966 hasOpenAccess W918148966 @default.
- W918148966 hasPrimaryLocation W9181489661 @default.
- W918148966 hasRelatedWork W142129510 @default.
- W918148966 hasRelatedWork W142450822 @default.
- W918148966 hasRelatedWork W1430782836 @default.
- W918148966 hasRelatedWork W1489268493 @default.
- W918148966 hasRelatedWork W2036173094 @default.
- W918148966 hasRelatedWork W2099027195 @default.
- W918148966 hasRelatedWork W2148721218 @default.
- W918148966 hasRelatedWork W2155107551 @default.
- W918148966 hasRelatedWork W2262276186 @default.
- W918148966 hasRelatedWork W2355052100 @default.
- W918148966 hasRelatedWork W2616916366 @default.
- W918148966 hasRelatedWork W2937070862 @default.
- W918148966 hasRelatedWork W2950945709 @default.
- W918148966 hasRelatedWork W2962868099 @default.
- W918148966 hasRelatedWork W3010178847 @default.
- W918148966 hasRelatedWork W3011228690 @default.
- W918148966 hasRelatedWork W3110735768 @default.