Matches in SemOpenAlex for { <https://semopenalex.org/work/W3037890969> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W3037890969 abstract "Motivated by a wide variety of applications, ranging from stochastic optimization to dimension reduction through variable selection, the problem of estimating gradients accurately is of crucial importance in statistics and learning theory. We consider here the classic regression setup, where a real valued square integrable r.v. $Y$ is to be predicted upon observing a (possibly high dimensional) random vector $X$ by means of a predictive function $f(X)$ as accurately as possible in the mean-squared sense and study a nearest-neighbour-based pointwise estimate of the gradient of the optimal predictive function, the regression function $m(x)=mathbb{E}[Ymid X=x]$. Under classic smoothness conditions combined with the assumption that the tails of $Y-m(X)$ are sub-Gaussian, we prove nonasymptotic bounds improving upon those obtained for alternative estimation methods. Beyond the novel theoretical results established, several illustrative numerical experiments have been carried out. The latter provide strong empirical evidence that the estimation method proposed works very well for various statistical problems involving gradient estimation, namely dimensionality reduction, stochastic gradient descent optimization and quantifying disentanglement." @default.
- W3037890969 created "2020-07-02" @default.
- W3037890969 creator A5044726008 @default.
- W3037890969 creator A5082182482 @default.
- W3037890969 creator A5088738795 @default.
- W3037890969 date "2021-01-01" @default.
- W3037890969 modified "2023-10-18" @default.
- W3037890969 title "Nearest Neighbour Based Estimates of Gradients: Sharp Nonasymptotic Bounds and Applications" @default.
- W3037890969 cites W1481221946 @default.
- W3037890969 cites W1484867920 @default.
- W3037890969 cites W1509689762 @default.
- W3037890969 cites W1523985187 @default.
- W3037890969 cites W1564947197 @default.
- W3037890969 cites W1597926580 @default.
- W3037890969 cites W1975547855 @default.
- W3037890969 cites W1978930294 @default.
- W3037890969 cites W1989164753 @default.
- W3037890969 cites W1997384244 @default.
- W3037890969 cites W2014497688 @default.
- W3037890969 cites W2022449465 @default.
- W3037890969 cites W2038845890 @default.
- W3037890969 cites W2046907549 @default.
- W3037890969 cites W2057032881 @default.
- W3037890969 cites W2065281378 @default.
- W3037890969 cites W2070991548 @default.
- W3037890969 cites W2080872718 @default.
- W3037890969 cites W2097462699 @default.
- W3037890969 cites W2097996627 @default.
- W3037890969 cites W2105608756 @default.
- W3037890969 cites W2144366961 @default.
- W3037890969 cites W2149066310 @default.
- W3037890969 cites W2166903773 @default.
- W3037890969 cites W2171050905 @default.
- W3037890969 cites W2411801785 @default.
- W3037890969 cites W2522078129 @default.
- W3037890969 cites W2740236269 @default.
- W3037890969 cites W2753738274 @default.
- W3037890969 cites W2899039999 @default.
- W3037890969 cites W2957135539 @default.
- W3037890969 cites W2963483869 @default.
- W3037890969 cites W2964122153 @default.
- W3037890969 cites W2970791107 @default.
- W3037890969 cites W2995540461 @default.
- W3037890969 cites W568673721 @default.
- W3037890969 cites W78182116 @default.
- W3037890969 hasPublicationYear "2021" @default.
- W3037890969 type Work @default.
- W3037890969 sameAs 3037890969 @default.
- W3037890969 citedByCount "0" @default.
- W3037890969 crossrefType "proceedings-article" @default.
- W3037890969 hasAuthorship W3037890969A5044726008 @default.
- W3037890969 hasAuthorship W3037890969A5082182482 @default.
- W3037890969 hasAuthorship W3037890969A5088738795 @default.
- W3037890969 hasBestOaLocation W30378909691 @default.
- W3037890969 hasConcept C11413529 @default.
- W3037890969 hasConcept C121332964 @default.
- W3037890969 hasConcept C121864883 @default.
- W3037890969 hasConcept C33923547 @default.
- W3037890969 hasConcept C41008148 @default.
- W3037890969 hasConceptScore W3037890969C11413529 @default.
- W3037890969 hasConceptScore W3037890969C121332964 @default.
- W3037890969 hasConceptScore W3037890969C121864883 @default.
- W3037890969 hasConceptScore W3037890969C33923547 @default.
- W3037890969 hasConceptScore W3037890969C41008148 @default.
- W3037890969 hasLocation W30378909691 @default.
- W3037890969 hasLocation W30378909692 @default.
- W3037890969 hasOpenAccess W3037890969 @default.
- W3037890969 hasPrimaryLocation W30378909691 @default.
- W3037890969 hasRelatedWork W1979597421 @default.
- W3037890969 hasRelatedWork W2007980826 @default.
- W3037890969 hasRelatedWork W2061531152 @default.
- W3037890969 hasRelatedWork W2077600819 @default.
- W3037890969 hasRelatedWork W2386767533 @default.
- W3037890969 hasRelatedWork W2748952813 @default.
- W3037890969 hasRelatedWork W2899084033 @default.
- W3037890969 hasRelatedWork W3002753104 @default.
- W3037890969 hasRelatedWork W4225152035 @default.
- W3037890969 hasRelatedWork W4245490552 @default.
- W3037890969 isParatext "false" @default.
- W3037890969 isRetracted "false" @default.
- W3037890969 magId "3037890969" @default.
- W3037890969 workType "article" @default.