Matches in SemOpenAlex for { <https://semopenalex.org/work/W2147118250> ?p ?o ?g. }
- W2147118250 abstract "We consider the problem of improving the efficiency of randomized Fourier feature maps to accelerate training and testing speed of kernel methods on large datasets. These approximate feature maps arise as Monte Carlo approximations to integral representations of shift-invariant kernel functions (e.g., Gaussian kernel). In this paper, we propose to use Quasi-Monte Carlo (QMC) approximations instead, where the relevant integrands are evaluated on a low-discrepancy sequence of points as opposed to random point sets as in the Monte Carlo approach. We derive a new discrepancy measure called box discrepancy based on theoretical characterizations of the integration error with respect to a given sequence. We then propose to learn QMC sequences adapted to our setting based on explicit box discrepancy minimization. Our theoretical analyses are complemented with empirical results that demonstrate the effectiveness of classical and adaptive QMC techniques for this problem." @default.
- W2147118250 created "2016-06-24" @default.
- W2147118250 creator A5033006662 @default.
- W2147118250 creator A5033659558 @default.
- W2147118250 creator A5058073838 @default.
- W2147118250 creator A5072151198 @default.
- W2147118250 date "2014-12-29" @default.
- W2147118250 modified "2023-09-27" @default.
- W2147118250 title "Quasi-Monte Carlo Feature Maps for Shift-Invariant Kernels" @default.
- W2147118250 cites W1751437809 @default.
- W2147118250 cites W1883420250 @default.
- W2147118250 cites W1946137962 @default.
- W2147118250 cites W1971922216 @default.
- W2147118250 cites W1978501336 @default.
- W2147118250 cites W1981743146 @default.
- W2147118250 cites W1988813039 @default.
- W2147118250 cites W1990470142 @default.
- W2147118250 cites W1997865285 @default.
- W2147118250 cites W2043419861 @default.
- W2147118250 cites W2047278710 @default.
- W2147118250 cites W2069618050 @default.
- W2147118250 cites W2086092605 @default.
- W2147118250 cites W2104290444 @default.
- W2147118250 cites W2105527258 @default.
- W2147118250 cites W2106149169 @default.
- W2147118250 cites W2113642685 @default.
- W2147118250 cites W2118563516 @default.
- W2147118250 cites W2119471399 @default.
- W2147118250 cites W2123602290 @default.
- W2147118250 cites W2123606421 @default.
- W2147118250 cites W2124331852 @default.
- W2147118250 cites W2138153286 @default.
- W2147118250 cites W2144902422 @default.
- W2147118250 cites W2146766088 @default.
- W2147118250 cites W2146897752 @default.
- W2147118250 cites W2151084831 @default.
- W2147118250 cites W2151683004 @default.
- W2147118250 cites W2163255905 @default.
- W2147118250 cites W2418462635 @default.
- W2147118250 cites W2949554124 @default.
- W2147118250 cites W2952772488 @default.
- W2147118250 cites W2952853356 @default.
- W2147118250 cites W2963942108 @default.
- W2147118250 cites W2964265625 @default.
- W2147118250 hasPublicationYear "2014" @default.
- W2147118250 type Work @default.
- W2147118250 sameAs 2147118250 @default.
- W2147118250 citedByCount "37" @default.
- W2147118250 countsByYear W21471182502014 @default.
- W2147118250 countsByYear W21471182502015 @default.
- W2147118250 countsByYear W21471182502016 @default.
- W2147118250 countsByYear W21471182502017 @default.
- W2147118250 countsByYear W21471182502018 @default.
- W2147118250 countsByYear W21471182502019 @default.
- W2147118250 countsByYear W21471182502020 @default.
- W2147118250 countsByYear W21471182502021 @default.
- W2147118250 crossrefType "posted-content" @default.
- W2147118250 hasAuthorship W2147118250A5033006662 @default.
- W2147118250 hasAuthorship W2147118250A5033659558 @default.
- W2147118250 hasAuthorship W2147118250A5058073838 @default.
- W2147118250 hasAuthorship W2147118250A5072151198 @default.
- W2147118250 hasConcept C105795698 @default.
- W2147118250 hasConcept C111350023 @default.
- W2147118250 hasConcept C11413529 @default.
- W2147118250 hasConcept C118615104 @default.
- W2147118250 hasConcept C121332964 @default.
- W2147118250 hasConcept C122280245 @default.
- W2147118250 hasConcept C12267149 @default.
- W2147118250 hasConcept C126255220 @default.
- W2147118250 hasConcept C13153151 @default.
- W2147118250 hasConcept C132725507 @default.
- W2147118250 hasConcept C138885662 @default.
- W2147118250 hasConcept C154945302 @default.
- W2147118250 hasConcept C163716315 @default.
- W2147118250 hasConcept C190470478 @default.
- W2147118250 hasConcept C19499675 @default.
- W2147118250 hasConcept C2776401178 @default.
- W2147118250 hasConcept C28826006 @default.
- W2147118250 hasConcept C33923547 @default.
- W2147118250 hasConcept C37914503 @default.
- W2147118250 hasConcept C41008148 @default.
- W2147118250 hasConcept C41895202 @default.
- W2147118250 hasConcept C62520636 @default.
- W2147118250 hasConcept C63320529 @default.
- W2147118250 hasConcept C74193536 @default.
- W2147118250 hasConceptScore W2147118250C105795698 @default.
- W2147118250 hasConceptScore W2147118250C111350023 @default.
- W2147118250 hasConceptScore W2147118250C11413529 @default.
- W2147118250 hasConceptScore W2147118250C118615104 @default.
- W2147118250 hasConceptScore W2147118250C121332964 @default.
- W2147118250 hasConceptScore W2147118250C122280245 @default.
- W2147118250 hasConceptScore W2147118250C12267149 @default.
- W2147118250 hasConceptScore W2147118250C126255220 @default.
- W2147118250 hasConceptScore W2147118250C13153151 @default.
- W2147118250 hasConceptScore W2147118250C132725507 @default.
- W2147118250 hasConceptScore W2147118250C138885662 @default.
- W2147118250 hasConceptScore W2147118250C154945302 @default.
- W2147118250 hasConceptScore W2147118250C163716315 @default.
- W2147118250 hasConceptScore W2147118250C190470478 @default.
- W2147118250 hasConceptScore W2147118250C19499675 @default.