Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285509650> ?p ?o ?g. }
- W4285509650 endingPage "1295" @default.
- W4285509650 startingPage "1282" @default.
- W4285509650 abstract "Outliers widely occur in big-data applications and may severely affect statistical estimation and inference. In this article, a framework of outlier-resistant estimation is introduced to robustify an arbitrarily given loss function. It has a close connection to the method of trimming and includes explicit outlyingness parameters for all samples, which in turn facilitates computation, theory, and parameter tuning. To tackle the issues of nonconvexity and nonsmoothness, we develop scalable algorithms with implementation ease and guaranteed fast convergence. In particular, a new technique is proposed to alleviate the requirement on the starting point such that on regular datasets, the number of data resamplings can be substantially reduced. Based on combined statistical and computational treatments, we are able to perform nonasymptotic analysis beyond M-estimation. The obtained resistant estimators, though not necessarily globally or even locally optimal, enjoy minimax rate optimality in both low dimensions and high dimensions. Experiments in regression, classification, and neural networks show excellent performance of the proposed methodology at the occurrence of gross outliers. Supplementary materials for this article are available online." @default.
- W4285509650 created "2022-07-15" @default.
- W4285509650 creator A5035282379 @default.
- W4285509650 creator A5040391223 @default.
- W4285509650 creator A5043540428 @default.
- W4285509650 date "2021-01-14" @default.
- W4285509650 modified "2023-10-01" @default.
- W4285509650 title "Gaining Outlier Resistance With Progressive Quantiles: Fast Algorithms and Theoretical Studies" @default.
- W4285509650 cites W1460189015 @default.
- W4285509650 cites W1485306023 @default.
- W4285509650 cites W1523144742 @default.
- W4285509650 cites W1523985187 @default.
- W4285509650 cites W1969515697 @default.
- W4285509650 cites W1978153422 @default.
- W4285509650 cites W1997859853 @default.
- W4285509650 cites W2013527206 @default.
- W4285509650 cites W2029888153 @default.
- W4285509650 cites W2033468335 @default.
- W4285509650 cites W2034078604 @default.
- W4285509650 cites W2037030821 @default.
- W4285509650 cites W2037760741 @default.
- W4285509650 cites W2038020460 @default.
- W4285509650 cites W2039385587 @default.
- W4285509650 cites W2039943102 @default.
- W4285509650 cites W2073604453 @default.
- W4285509650 cites W2088883866 @default.
- W4285509650 cites W2092058109 @default.
- W4285509650 cites W2129131372 @default.
- W4285509650 cites W2135046866 @default.
- W4285509650 cites W2146225536 @default.
- W4285509650 cites W2147898188 @default.
- W4285509650 cites W2149414429 @default.
- W4285509650 cites W2155524176 @default.
- W4285509650 cites W2159384462 @default.
- W4285509650 cites W2289917018 @default.
- W4285509650 cites W2602934021 @default.
- W4285509650 cites W2617337526 @default.
- W4285509650 cites W2790297619 @default.
- W4285509650 cites W2898542396 @default.
- W4285509650 cites W2963172671 @default.
- W4285509650 cites W2963826549 @default.
- W4285509650 cites W2963927498 @default.
- W4285509650 cites W2990681736 @default.
- W4285509650 cites W3101094059 @default.
- W4285509650 cites W3105728206 @default.
- W4285509650 cites W3125188740 @default.
- W4285509650 cites W36711357 @default.
- W4285509650 cites W4200303272 @default.
- W4285509650 cites W4205806204 @default.
- W4285509650 cites W4250589301 @default.
- W4285509650 cites W4255230573 @default.
- W4285509650 doi "https://doi.org/10.1080/01621459.2020.1850460" @default.
- W4285509650 hasPublicationYear "2021" @default.
- W4285509650 type Work @default.
- W4285509650 citedByCount "4" @default.
- W4285509650 countsByYear W42855096502023 @default.
- W4285509650 crossrefType "journal-article" @default.
- W4285509650 hasAuthorship W4285509650A5035282379 @default.
- W4285509650 hasAuthorship W4285509650A5040391223 @default.
- W4285509650 hasAuthorship W4285509650A5043540428 @default.
- W4285509650 hasBestOaLocation W42855096502 @default.
- W4285509650 hasConcept C105795698 @default.
- W4285509650 hasConcept C111919701 @default.
- W4285509650 hasConcept C11413529 @default.
- W4285509650 hasConcept C118671147 @default.
- W4285509650 hasConcept C124101348 @default.
- W4285509650 hasConcept C126255220 @default.
- W4285509650 hasConcept C149728462 @default.
- W4285509650 hasConcept C149782125 @default.
- W4285509650 hasConcept C154945302 @default.
- W4285509650 hasConcept C162324750 @default.
- W4285509650 hasConcept C185429906 @default.
- W4285509650 hasConcept C2776214188 @default.
- W4285509650 hasConcept C2777303404 @default.
- W4285509650 hasConcept C33923547 @default.
- W4285509650 hasConcept C41008148 @default.
- W4285509650 hasConcept C45374587 @default.
- W4285509650 hasConcept C48044578 @default.
- W4285509650 hasConcept C50522688 @default.
- W4285509650 hasConcept C56951928 @default.
- W4285509650 hasConcept C67226441 @default.
- W4285509650 hasConcept C77088390 @default.
- W4285509650 hasConcept C79337645 @default.
- W4285509650 hasConceptScore W4285509650C105795698 @default.
- W4285509650 hasConceptScore W4285509650C111919701 @default.
- W4285509650 hasConceptScore W4285509650C11413529 @default.
- W4285509650 hasConceptScore W4285509650C118671147 @default.
- W4285509650 hasConceptScore W4285509650C124101348 @default.
- W4285509650 hasConceptScore W4285509650C126255220 @default.
- W4285509650 hasConceptScore W4285509650C149728462 @default.
- W4285509650 hasConceptScore W4285509650C149782125 @default.
- W4285509650 hasConceptScore W4285509650C154945302 @default.
- W4285509650 hasConceptScore W4285509650C162324750 @default.
- W4285509650 hasConceptScore W4285509650C185429906 @default.
- W4285509650 hasConceptScore W4285509650C2776214188 @default.
- W4285509650 hasConceptScore W4285509650C2777303404 @default.
- W4285509650 hasConceptScore W4285509650C33923547 @default.
- W4285509650 hasConceptScore W4285509650C41008148 @default.