FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3036291575> ?p ?o ?g. }

• W3036291575 abstract "The goal of compressed sensing is to estimate a high dimensional vector from an underdetermined system of noisy linear equations. In analogy to classical compressed sensing, here we assume a generative model as a prior, that is, we assume the vector is represented by a deep generative model $G: mathbb{R}^k rightarrow mathbb{R}^n$. Classical recovery approaches such as empirical risk minimization (ERM) are guaranteed to succeed when the measurement matrix is sub-Gaussian. However, when the measurement matrix and measurements are heavy-tailed or have outliers, recovery may fail dramatically. In this paper we propose an algorithm inspired by the Median-of-Means (MOM). Our algorithm guarantees recovery for heavy-tailed data, even in the presence of outliers. Theoretically, our results show our novel MOM-based algorithm enjoys the same sample complexity guarantees as ERM under sub-Gaussian assumptions. Our experiments validate both aspects of our claims: other algorithms are indeed fragile and fail under heavy-tailed and/or corrupted data, while our approach exhibits the predicted robustness." @default.
• W3036291575 created "2020-06-25" @default.
• W3036291575 creator A5004075655 @default.
• W3036291575 creator A5014141359 @default.
• W3036291575 creator A5053978837 @default.
• W3036291575 creator A5083137649 @default.
• W3036291575 date "2020-06-16" @default.
• W3036291575 modified "2023-09-27" @default.
• W3036291575 title "Robust compressed sensing of generative models." @default.
• W3036291575 cites W1522301498 @default.
• W3036291575 cites W1523985187 @default.
• W3036291575 cites W1684009677 @default.
• W3036291575 cites W1986736933 @default.
• W3036291575 cites W2015418199 @default.
• W3036291575 cites W2045385910 @default.
• W3036291575 cites W2080745194 @default.
• W3036291575 cites W2089493818 @default.
• W3036291575 cites W2097474301 @default.
• W3036291575 cites W2098316308 @default.
• W3036291575 cites W2104266187 @default.
• W3036291575 cites W2107179775 @default.
• W3036291575 cites W2112796928 @default.
• W3036291575 cites W2127271355 @default.
• W3036291575 cites W2135046866 @default.
• W3036291575 cites W2145096794 @default.
• W3036291575 cites W2164452299 @default.
• W3036291575 cites W2173520492 @default.
• W3036291575 cites W2296616510 @default.
• W3036291575 cites W2480397354 @default.
• W3036291575 cites W2488280448 @default.
• W3036291575 cites W2566240941 @default.
• W3036291575 cites W2614254310 @default.
• W3036291575 cites W2617427826 @default.
• W3036291575 cites W2625429595 @default.
• W3036291575 cites W2625885504 @default.
• W3036291575 cites W2723734710 @default.
• W3036291575 cites W2766527293 @default.
• W3036291575 cites W2769179713 @default.
• W3036291575 cites W2785446525 @default.
• W3036291575 cites W2794168668 @default.
• W3036291575 cites W2806196585 @default.
• W3036291575 cites W2810458093 @default.
• W3036291575 cites W2883105305 @default.
• W3036291575 cites W2898001268 @default.
• W3036291575 cites W2909813108 @default.
• W3036291575 cites W2922064308 @default.
• W3036291575 cites W2946940238 @default.
• W3036291575 cites W2947708079 @default.
• W3036291575 cites W2949521215 @default.
• W3036291575 cites W2949536516 @default.
• W3036291575 cites W2951004968 @default.
• W3036291575 cites W2951227519 @default.
• W3036291575 cites W2952139899 @default.
• W3036291575 cites W2962909343 @default.
• W3036291575 cites W2963031612 @default.
• W3036291575 cites W2963719122 @default.
• W3036291575 cites W2963927498 @default.
• W3036291575 cites W2964135521 @default.
• W3036291575 cites W2966991718 @default.
• W3036291575 cites W2967227187 @default.
• W3036291575 cites W2970109978 @default.
• W3036291575 cites W2970877313 @default.
• W3036291575 cites W2971117199 @default.
• W3036291575 cites W2982750369 @default.
• W3036291575 cites W2987357275 @default.
• W3036291575 cites W2993730048 @default.
• W3036291575 cites W2994196995 @default.
• W3036291575 cites W2994907479 @default.
• W3036291575 cites W3005530837 @default.
• W3036291575 cites W3017335724 @default.
• W3036291575 cites W3022022653 @default.
• W3036291575 cites W3025242627 @default.
• W3036291575 cites W3033846281 @default.
• W3036291575 cites W3034279703 @default.
• W3036291575 cites W3111890340 @default.
• W3036291575 cites W2821662226 @default.
• W3036291575 cites W2945068291 @default.
• W3036291575 hasPublicationYear "2020" @default.
• W3036291575 type Work @default.
• W3036291575 sameAs 3036291575 @default.
• W3036291575 citedByCount "4" @default.
• W3036291575 countsByYear W30362915752020 @default.
• W3036291575 countsByYear W30362915752021 @default.
• W3036291575 crossrefType "posted-content" @default.
• W3036291575 hasAuthorship W3036291575A5004075655 @default.
• W3036291575 hasAuthorship W3036291575A5014141359 @default.
• W3036291575 hasAuthorship W3036291575A5053978837 @default.
• W3036291575 hasAuthorship W3036291575A5083137649 @default.
• W3036291575 hasConcept C104317684 @default.
• W3036291575 hasConcept C106487976 @default.
• W3036291575 hasConcept C11413529 @default.
• W3036291575 hasConcept C121332964 @default.
• W3036291575 hasConcept C124851039 @default.
• W3036291575 hasConcept C126255220 @default.
• W3036291575 hasConcept C147764199 @default.
• W3036291575 hasConcept C154945302 @default.
• W3036291575 hasConcept C159985019 @default.
• W3036291575 hasConcept C163716315 @default.
• W3036291575 hasConcept C167966045 @default.
• W3036291575 hasConcept C179690561 @default.

• next