Matches in SemOpenAlex for { <https://semopenalex.org/work/W4284963688> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4284963688 endingPage "109266" @default.
- W4284963688 startingPage "109266" @default.
- W4284963688 abstract "Supervised learning is an important tool for data mining and knowledge discovery. The hyper-parameter in learning models usually has a significant impact on the generalization performance of supervised learning model. Although some state-of-the-art hyper-parameter optimizers, such as cross-validation (CV) and its improvements, have been widely used in many applications, there still exist some limitations including the low efficiency and the error variation from data partition. To address these issues, we propose the sign similarity to distinguish between over-fitting and under-fitting in supervised learning. On this basis, the minimal symmetric similarity criterion (MSSC) is proposed to optimize hyper-parameters. It provides an equivalent condition of well-fitting, i.e., the well-fitted hyper-parameter should have a minimal symmetric similarity. Compared with CV, the criterion is more efficient and could avoid error variation as the symmetric similarity is based on the whole data set without partition. The reasonableness of the proposed MSSC is proved for well-posed learning problems. Also, corresponding assumptions are explained by means of theoretical analysis and empirical verification. Both theoretical analysis and experimental results indicate that the efficiency of MSSC is significantly higher than popular optimizers like 10-CV. Experimental results demonstrate that MSSC is comparable with or outperforms the state-of-the-art optimizers in three kinds of learning tasks from the statistical perspective." @default.
- W4284963688 created "2022-07-10" @default.
- W4284963688 creator A5012129858 @default.
- W4284963688 creator A5031734464 @default.
- W4284963688 creator A5033746570 @default.
- W4284963688 creator A5056860818 @default.
- W4284963688 date "2022-09-01" @default.
- W4284963688 modified "2023-10-01" @default.
- W4284963688 title "An efficient hyper-parameter optimization method for supervised learning" @default.
- W4284963688 cites W1982405594 @default.
- W4284963688 cites W1986543815 @default.
- W4284963688 cites W1993631132 @default.
- W4284963688 cites W1996020380 @default.
- W4284963688 cites W2017848200 @default.
- W4284963688 cites W2019635781 @default.
- W4284963688 cites W2023380651 @default.
- W4284963688 cites W2073792037 @default.
- W4284963688 cites W2140229565 @default.
- W4284963688 cites W2290300505 @default.
- W4284963688 cites W2358876993 @default.
- W4284963688 cites W2510725918 @default.
- W4284963688 cites W2516499113 @default.
- W4284963688 cites W2525261790 @default.
- W4284963688 cites W2605548712 @default.
- W4284963688 cites W2749959226 @default.
- W4284963688 cites W2808275179 @default.
- W4284963688 cites W2942137712 @default.
- W4284963688 cites W3043356793 @default.
- W4284963688 cites W3092969547 @default.
- W4284963688 cites W3130777809 @default.
- W4284963688 cites W4210675695 @default.
- W4284963688 cites W3146806684 @default.
- W4284963688 doi "https://doi.org/10.1016/j.asoc.2022.109266" @default.
- W4284963688 hasPublicationYear "2022" @default.
- W4284963688 type Work @default.
- W4284963688 citedByCount "3" @default.
- W4284963688 countsByYear W42849636882023 @default.
- W4284963688 crossrefType "journal-article" @default.
- W4284963688 hasAuthorship W4284963688A5012129858 @default.
- W4284963688 hasAuthorship W4284963688A5031734464 @default.
- W4284963688 hasAuthorship W4284963688A5033746570 @default.
- W4284963688 hasAuthorship W4284963688A5056860818 @default.
- W4284963688 hasConcept C103278499 @default.
- W4284963688 hasConcept C11413529 @default.
- W4284963688 hasConcept C114614502 @default.
- W4284963688 hasConcept C115961682 @default.
- W4284963688 hasConcept C117765406 @default.
- W4284963688 hasConcept C119857082 @default.
- W4284963688 hasConcept C124101348 @default.
- W4284963688 hasConcept C12713177 @default.
- W4284963688 hasConcept C134306372 @default.
- W4284963688 hasConcept C136389625 @default.
- W4284963688 hasConcept C153180895 @default.
- W4284963688 hasConcept C154945302 @default.
- W4284963688 hasConcept C177148314 @default.
- W4284963688 hasConcept C33923547 @default.
- W4284963688 hasConcept C41008148 @default.
- W4284963688 hasConcept C42812 @default.
- W4284963688 hasConcept C50644808 @default.
- W4284963688 hasConcept C8038995 @default.
- W4284963688 hasConceptScore W4284963688C103278499 @default.
- W4284963688 hasConceptScore W4284963688C11413529 @default.
- W4284963688 hasConceptScore W4284963688C114614502 @default.
- W4284963688 hasConceptScore W4284963688C115961682 @default.
- W4284963688 hasConceptScore W4284963688C117765406 @default.
- W4284963688 hasConceptScore W4284963688C119857082 @default.
- W4284963688 hasConceptScore W4284963688C124101348 @default.
- W4284963688 hasConceptScore W4284963688C12713177 @default.
- W4284963688 hasConceptScore W4284963688C134306372 @default.
- W4284963688 hasConceptScore W4284963688C136389625 @default.
- W4284963688 hasConceptScore W4284963688C153180895 @default.
- W4284963688 hasConceptScore W4284963688C154945302 @default.
- W4284963688 hasConceptScore W4284963688C177148314 @default.
- W4284963688 hasConceptScore W4284963688C33923547 @default.
- W4284963688 hasConceptScore W4284963688C41008148 @default.
- W4284963688 hasConceptScore W4284963688C42812 @default.
- W4284963688 hasConceptScore W4284963688C50644808 @default.
- W4284963688 hasConceptScore W4284963688C8038995 @default.
- W4284963688 hasLocation W42849636881 @default.
- W4284963688 hasOpenAccess W4284963688 @default.
- W4284963688 hasPrimaryLocation W42849636881 @default.
- W4284963688 hasRelatedWork W2953328427 @default.
- W4284963688 hasRelatedWork W2971361125 @default.
- W4284963688 hasRelatedWork W3046775127 @default.
- W4284963688 hasRelatedWork W3091943846 @default.
- W4284963688 hasRelatedWork W3094076422 @default.
- W4284963688 hasRelatedWork W3162567751 @default.
- W4284963688 hasRelatedWork W4220686584 @default.
- W4284963688 hasRelatedWork W4221088574 @default.
- W4284963688 hasRelatedWork W4285260836 @default.
- W4284963688 hasRelatedWork W4319309271 @default.
- W4284963688 hasVolume "126" @default.
- W4284963688 isParatext "false" @default.
- W4284963688 isRetracted "false" @default.
- W4284963688 workType "article" @default.