Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048604853> ?p ?o ?g. }
- W3048604853 abstract "Deep networks have been used to learn transferable representations for domain adaptation. Existing deep domain adaptation methods systematically employ popular hand-crafted networks designed specifically for image-classification tasks, leading to sub-optimal domain adaptation performance. In this paper, we present Neural Architecture Search for Domain Adaptation (NASDA), a principle framework that leverages differentiable neural architecture search to derive the optimal network architecture for domain adaptation task. NASDA is designed with two novel training strategies: neural architecture search with multi-kernel Maximum Mean Discrepancy to derive the optimal architecture, and adversarial training between a feature generator and a batch of classifiers to consolidate the feature generator. We demonstrate experimentally that NASDA leads to state-of-the-art performance on several domain adaptation benchmarks." @default.
- W3048604853 created "2020-08-18" @default.
- W3048604853 creator A5067025277 @default.
- W3048604853 creator A5068469395 @default.
- W3048604853 date "2020-08-13" @default.
- W3048604853 modified "2023-10-14" @default.
- W3048604853 title "Network Architecture Search for Domain Adaptation" @default.
- W3048604853 cites W1565327149 @default.
- W3048604853 cites W1569788011 @default.
- W3048604853 cites W16455933 @default.
- W3048604853 cites W2090411045 @default.
- W3048604853 cites W2097117768 @default.
- W3048604853 cites W2104094955 @default.
- W3048604853 cites W2105523772 @default.
- W3048604853 cites W2117876524 @default.
- W3048604853 cites W2118858186 @default.
- W3048604853 cites W2122084318 @default.
- W3048604853 cites W2125865219 @default.
- W3048604853 cites W2148440006 @default.
- W3048604853 cites W2159291411 @default.
- W3048604853 cites W2162651021 @default.
- W3048604853 cites W2194775991 @default.
- W3048604853 cites W2212660284 @default.
- W3048604853 cites W2266822037 @default.
- W3048604853 cites W2270245739 @default.
- W3048604853 cites W2511131004 @default.
- W3048604853 cites W2593768305 @default.
- W3048604853 cites W2598581049 @default.
- W3048604853 cites W2604763608 @default.
- W3048604853 cites W2605488490 @default.
- W3048604853 cites W2617027347 @default.
- W3048604853 cites W2741430497 @default.
- W3048604853 cites W2744915377 @default.
- W3048604853 cites W2751836095 @default.
- W3048604853 cites W2767722847 @default.
- W3048604853 cites W2786559811 @default.
- W3048604853 cites W2792485748 @default.
- W3048604853 cites W2795155917 @default.
- W3048604853 cites W2804529069 @default.
- W3048604853 cites W2810857565 @default.
- W3048604853 cites W2883188678 @default.
- W3048604853 cites W2885820039 @default.
- W3048604853 cites W2890336707 @default.
- W3048604853 cites W2891778567 @default.
- W3048604853 cites W2905672847 @default.
- W3048604853 cites W2913732044 @default.
- W3048604853 cites W2945979290 @default.
- W3048604853 cites W2950240247 @default.
- W3048604853 cites W2950635901 @default.
- W3048604853 cites W2955425717 @default.
- W3048604853 cites W2962687275 @default.
- W3048604853 cites W2962740476 @default.
- W3048604853 cites W2962746461 @default.
- W3048604853 cites W2962793481 @default.
- W3048604853 cites W2962808524 @default.
- W3048604853 cites W2962835968 @default.
- W3048604853 cites W2962892300 @default.
- W3048604853 cites W2963137684 @default.
- W3048604853 cites W2963168418 @default.
- W3048604853 cites W2963233958 @default.
- W3048604853 cites W2963327260 @default.
- W3048604853 cites W2963374479 @default.
- W3048604853 cites W2963444790 @default.
- W3048604853 cites W2963474950 @default.
- W3048604853 cites W2963506806 @default.
- W3048604853 cites W2963536136 @default.
- W3048604853 cites W2963778169 @default.
- W3048604853 cites W2963784072 @default.
- W3048604853 cites W2963821229 @default.
- W3048604853 cites W2963826681 @default.
- W3048604853 cites W2963918968 @default.
- W3048604853 cites W2964024268 @default.
- W3048604853 cites W2964029277 @default.
- W3048604853 cites W2964081403 @default.
- W3048604853 cites W2964081807 @default.
- W3048604853 cites W2964259004 @default.
- W3048604853 cites W2964266860 @default.
- W3048604853 cites W2964278684 @default.
- W3048604853 cites W2964288524 @default.
- W3048604853 cites W2964294659 @default.
- W3048604853 cites W2965658867 @default.
- W3048604853 cites W2965954137 @default.
- W3048604853 cites W2981720610 @default.
- W3048604853 cites W2981748264 @default.
- W3048604853 cites W2990010862 @default.
- W3048604853 cites W2997310315 @default.
- W3048604853 cites W3118608800 @default.
- W3048604853 cites W2904817185 @default.
- W3048604853 doi "https://doi.org/10.48550/arxiv.2008.05706" @default.
- W3048604853 hasPublicationYear "2020" @default.
- W3048604853 type Work @default.
- W3048604853 sameAs 3048604853 @default.
- W3048604853 citedByCount "2" @default.
- W3048604853 countsByYear W30486048532021 @default.
- W3048604853 crossrefType "posted-content" @default.
- W3048604853 hasAuthorship W3048604853A5067025277 @default.
- W3048604853 hasAuthorship W3048604853A5068469395 @default.
- W3048604853 hasBestOaLocation W30486048531 @default.
- W3048604853 hasConcept C119857082 @default.
- W3048604853 hasConcept C120665830 @default.