Matches in SemOpenAlex for { <https://semopenalex.org/work/W3135514384> ?p ?o ?g. }
- W3135514384 abstract "Conditional quantile estimation is a key statistical learning challenge motivated by the need to quantify uncertainty in predictions or to model a diverse population without being overly reductive. As such, many models have been developed for this problem. Adopting a meta viewpoint, we propose a general framework (inspired by neural network optimization) for aggregating any number of conditional quantile models in order to boost predictive accuracy. We consider weighted ensembling strategies of increasing flexibility where the weights may vary over individual models, quantile levels, and feature values. An appeal of our approach is its portability: we ensure that estimated quantiles at adjacent levels do not cross by applying simple transformations through which gradients can be backpropagated, and this allows us to leverage the modern deep learning toolkit for building quantile ensembles. Our experiments confirm that ensembling can lead to big gains in accuracy, even when the constituent models are themselves powerful and flexible." @default.
- W3135514384 created "2021-03-15" @default.
- W3135514384 creator A5000245150 @default.
- W3135514384 creator A5010753342 @default.
- W3135514384 creator A5065728469 @default.
- W3135514384 creator A5074303284 @default.
- W3135514384 creator A5076708141 @default.
- W3135514384 date "2021-02-26" @default.
- W3135514384 modified "2023-09-27" @default.
- W3135514384 title "Deep Quantile Aggregation." @default.
- W3135514384 cites W1534477342 @default.
- W3135514384 cites W1573647811 @default.
- W3135514384 cites W1582036668 @default.
- W3135514384 cites W1803250861 @default.
- W3135514384 cites W1965755064 @default.
- W3135514384 cites W1980264541 @default.
- W3135514384 cites W1991452110 @default.
- W3135514384 cites W2008709824 @default.
- W3135514384 cites W2025720061 @default.
- W3135514384 cites W2043462999 @default.
- W3135514384 cites W2047445560 @default.
- W3135514384 cites W2053589029 @default.
- W3135514384 cites W2056132907 @default.
- W3135514384 cites W2057389198 @default.
- W3135514384 cites W2065861083 @default.
- W3135514384 cites W2066035378 @default.
- W3135514384 cites W2095015602 @default.
- W3135514384 cites W2111018099 @default.
- W3135514384 cites W2121798652 @default.
- W3135514384 cites W2132862423 @default.
- W3135514384 cites W2133479700 @default.
- W3135514384 cites W2133596638 @default.
- W3135514384 cites W2141484579 @default.
- W3135514384 cites W2145890287 @default.
- W3135514384 cites W2150884987 @default.
- W3135514384 cites W2158840489 @default.
- W3135514384 cites W2162859319 @default.
- W3135514384 cites W2165231911 @default.
- W3135514384 cites W2167250202 @default.
- W3135514384 cites W2229320441 @default.
- W3135514384 cites W2286305802 @default.
- W3135514384 cites W2460864273 @default.
- W3135514384 cites W2530641167 @default.
- W3135514384 cites W2599472182 @default.
- W3135514384 cites W2625224297 @default.
- W3135514384 cites W2761168002 @default.
- W3135514384 cites W2765302304 @default.
- W3135514384 cites W2768348081 @default.
- W3135514384 cites W2803408063 @default.
- W3135514384 cites W2810041204 @default.
- W3135514384 cites W2901746906 @default.
- W3135514384 cites W2921829327 @default.
- W3135514384 cites W2944327260 @default.
- W3135514384 cites W2946063682 @default.
- W3135514384 cites W2946130033 @default.
- W3135514384 cites W2963081736 @default.
- W3135514384 cites W2963238274 @default.
- W3135514384 cites W2963890379 @default.
- W3135514384 cites W2964121744 @default.
- W3135514384 cites W2964208955 @default.
- W3135514384 cites W2964308564 @default.
- W3135514384 cites W2970112944 @default.
- W3135514384 cites W2970971581 @default.
- W3135514384 cites W2971196730 @default.
- W3135514384 cites W2972895743 @default.
- W3135514384 cites W2997129641 @default.
- W3135514384 cites W3011307848 @default.
- W3135514384 cites W3017895332 @default.
- W3135514384 cites W3023659858 @default.
- W3135514384 cites W3033794115 @default.
- W3135514384 cites W3036188985 @default.
- W3135514384 cites W3044159891 @default.
- W3135514384 cites W3046998562 @default.
- W3135514384 cites W3091667719 @default.
- W3135514384 cites W3099796618 @default.
- W3135514384 cites W3100400430 @default.
- W3135514384 cites W3103465772 @default.
- W3135514384 cites W3120740533 @default.
- W3135514384 cites W3121494657 @default.
- W3135514384 cites W3124349228 @default.
- W3135514384 cites W3125857399 @default.
- W3135514384 cites W40442397 @default.
- W3135514384 cites W3148207383 @default.
- W3135514384 hasPublicationYear "2021" @default.
- W3135514384 type Work @default.
- W3135514384 sameAs 3135514384 @default.
- W3135514384 citedByCount "2" @default.
- W3135514384 countsByYear W31355143842021 @default.
- W3135514384 crossrefType "posted-content" @default.
- W3135514384 hasAuthorship W3135514384A5000245150 @default.
- W3135514384 hasAuthorship W3135514384A5010753342 @default.
- W3135514384 hasAuthorship W3135514384A5065728469 @default.
- W3135514384 hasAuthorship W3135514384A5074303284 @default.
- W3135514384 hasAuthorship W3135514384A5076708141 @default.
- W3135514384 hasConcept C105795698 @default.
- W3135514384 hasConcept C118671147 @default.
- W3135514384 hasConcept C119857082 @default.
- W3135514384 hasConcept C124101348 @default.
- W3135514384 hasConcept C13280743 @default.
- W3135514384 hasConcept C149782125 @default.