FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2567351292> ?p ?o ?g. }

• W2567351292 abstract "We describe a general technique that yields the first {em Statistical Query lower bounds} for a range of fundamental high-dimensional learning problems involving Gaussian distributions. Our main results are for the problems of (1) learning Gaussian mixture models (GMMs), and (2) robust (agnostic) learning of a single unknown Gaussian distribution. For each of these problems, we show a {em super-polynomial gap} between the (information-theoretic) sample complexity and the computational complexity of {em any} Statistical Query algorithm for the problem. Our SQ lower bound for Problem (1) is qualitatively matched by known learning algorithms for GMMs. Our lower bound for Problem (2) implies that the accuracy of the robust learning algorithm in~cite{DiakonikolasKKLMS16} is essentially best possible among all polynomial-time SQ algorithms. Our SQ lower bounds are attained via a unified moment-matching technique that is useful in other contexts and may be of broader interest. Our technique yields nearly-tight lower bounds for a number of related unsupervised estimation problems. Specifically, for the problems of (3) robust covariance estimation in spectral norm, and (4) robust sparse mean estimation, we establish a quadratic {em statistical--computational tradeoff} for SQ algorithms, matching known upper bounds. Finally, our technique can be used to obtain tight sample complexity lower bounds for high-dimensional {em testing} problems. Specifically, for the classical problem of robustly {em testing} an unknown mean (known covariance) Gaussian, our technique implies an information-theoretic sample lower bound that scales {em linearly} in the dimension. Our sample lower bound matches the sample complexity of the corresponding robust {em learning} problem and separates the sample complexity of robust testing from standard (non-robust) testing." @default.
• W2567351292 created "2017-01-06" @default.
• W2567351292 creator A5035962853 @default.
• W2567351292 creator A5054966819 @default.
• W2567351292 creator A5083699810 @default.
• W2567351292 date "2016-11-10" @default.
• W2567351292 modified "2023-09-27" @default.
• W2567351292 title "Statistical Query Lower Bounds for Robust Estimation of High-dimensional Gaussians and Gaussian Mixtures" @default.
• W2567351292 cites W1556014900 @default.
• W2567351292 cites W1574816920 @default.
• W2567351292 cites W1575031463 @default.
• W2567351292 cites W1581035799 @default.
• W2567351292 cites W1585566614 @default.
• W2567351292 cites W1626396758 @default.
• W2567351292 cites W1891181203 @default.
• W2567351292 cites W1956647075 @default.
• W2567351292 cites W1969015668 @default.
• W2567351292 cites W1976238508 @default.
• W2567351292 cites W1976697079 @default.
• W2567351292 cites W1985792472 @default.
• W2567351292 cites W1995897489 @default.
• W2567351292 cites W2014562510 @default.
• W2567351292 cites W2014565165 @default.
• W2567351292 cites W2019120673 @default.
• W2567351292 cites W2019363670 @default.
• W2567351292 cites W2026103826 @default.
• W2567351292 cites W2039892753 @default.
• W2567351292 cites W2046033161 @default.
• W2567351292 cites W2050498681 @default.
• W2567351292 cites W2051623559 @default.
• W2567351292 cites W2057214765 @default.
• W2567351292 cites W2062817469 @default.
• W2567351292 cites W2064680241 @default.
• W2567351292 cites W2064771454 @default.
• W2567351292 cites W2065563937 @default.
• W2567351292 cites W2072227363 @default.
• W2567351292 cites W2081874429 @default.
• W2567351292 cites W2095374884 @default.
• W2567351292 cites W2097714737 @default.
• W2567351292 cites W2104452301 @default.
• W2567351292 cites W2106458073 @default.
• W2567351292 cites W2109722477 @default.
• W2567351292 cites W2122170585 @default.
• W2567351292 cites W2133676097 @default.
• W2567351292 cites W2134169350 @default.
• W2567351292 cites W2140125566 @default.
• W2567351292 cites W2145615204 @default.
• W2567351292 cites W2146756121 @default.
• W2567351292 cites W2146852149 @default.
• W2567351292 cites W2163288162 @default.
• W2567351292 cites W2168002876 @default.
• W2567351292 cites W2178116779 @default.
• W2567351292 cites W2201177080 @default.
• W2567351292 cites W2210657914 @default.
• W2567351292 cites W2306288190 @default.
• W2567351292 cites W2336954923 @default.
• W2567351292 cites W2464464020 @default.
• W2567351292 cites W2517119954 @default.
• W2567351292 cites W2577046358 @default.
• W2567351292 cites W2592318711 @default.
• W2567351292 cites W2593349530 @default.
• W2567351292 cites W2949860624 @default.
• W2567351292 cites W2949977662 @default.
• W2567351292 cites W2950934461 @default.
• W2567351292 cites W2951177125 @default.
• W2567351292 cites W2962765673 @default.
• W2567351292 cites W2962960950 @default.
• W2567351292 cites W2963414662 @default.
• W2567351292 cites W2963595764 @default.
• W2567351292 cites W2963800155 @default.
• W2567351292 cites W2964317125 @default.
• W2567351292 cites W3104095169 @default.
• W2567351292 cites W3104815482 @default.
• W2567351292 cites W58673994 @default.
• W2567351292 cites W776622464 @default.
• W2567351292 cites W2166843037 @default.
• W2567351292 cites W2899169837 @default.
• W2567351292 hasPublicationYear "2016" @default.
• W2567351292 type Work @default.
• W2567351292 sameAs 2567351292 @default.
• W2567351292 citedByCount "13" @default.
• W2567351292 countsByYear W25673512922017 @default.
• W2567351292 countsByYear W25673512922018 @default.
• W2567351292 countsByYear W25673512922019 @default.
• W2567351292 countsByYear W25673512922021 @default.
• W2567351292 crossrefType "posted-content" @default.
• W2567351292 hasAuthorship W2567351292A5035962853 @default.
• W2567351292 hasAuthorship W2567351292A5054966819 @default.
• W2567351292 hasAuthorship W2567351292A5083699810 @default.
• W2567351292 hasConcept C105795698 @default.
• W2567351292 hasConcept C11413529 @default.
• W2567351292 hasConcept C114614502 @default.
• W2567351292 hasConcept C121332964 @default.
• W2567351292 hasConcept C129848803 @default.
• W2567351292 hasConcept C134306372 @default.
• W2567351292 hasConcept C158693339 @default.
• W2567351292 hasConcept C163716315 @default.
• W2567351292 hasConcept C165064840 @default.
• W2567351292 hasConcept C178650346 @default.
• W2567351292 hasConcept C179799912 @default.

• next