Matches in SemOpenAlex for { <https://semopenalex.org/work/W3144102935> ?p ?o ?g. }
- W3144102935 abstract "Unsupervised Domain Adaptation (UDA) transfers predictive models from a fully-labeled source domain to an unlabeled target domain. In some applications, however, it is expensive even to collect labels in the source domain, making most previous works impractical. To cope with this problem, recent work performed instance-wise cross-domain self-supervised learning, followed by an additional fine-tuning stage. However, the instance-wise self-supervised learning only learns and aligns low-level discriminative features. In this paper, we propose an end-to-end Prototypical Cross-domain Self-Supervised Learning (PCS) framework for Few-shot Unsupervised Domain Adaptation (FUDA). PCS not only performs cross-domain low-level feature alignment, but it also encodes and aligns semantic structures in the shared embedding space across domains. Our framework captures category-wise semantic structures of the data by in-domain prototypical contrastive learning; and performs feature alignment through cross-domain prototypical self-supervision. Compared with state-of-the-art methods, PCS improves the mean classification accuracy over different domain pairs on FUDA by 10.5%, 3.5%, 9.0%, and 13.2% on Office, Office-Home, VisDA-2017, and DomainNet, respectively. Our project page is at http://xyue.io/pcs-fuda/index.html" @default.
- W3144102935 created "2021-04-13" @default.
- W3144102935 creator A5007604342 @default.
- W3144102935 creator A5013030532 @default.
- W3144102935 creator A5027804169 @default.
- W3144102935 creator A5029105520 @default.
- W3144102935 creator A5047285420 @default.
- W3144102935 creator A5074250521 @default.
- W3144102935 creator A5078165161 @default.
- W3144102935 date "2021-03-30" @default.
- W3144102935 modified "2023-09-23" @default.
- W3144102935 title "Prototypical Cross-domain Self-supervised Learning for Few-shot Unsupervised Domain Adaptation" @default.
- W3144102935 cites W1565327149 @default.
- W3144102935 cites W1722318740 @default.
- W3144102935 cites W1731081199 @default.
- W3144102935 cites W1903029394 @default.
- W3144102935 cites W1920962657 @default.
- W3144102935 cites W2031342017 @default.
- W3144102935 cites W2099471712 @default.
- W3144102935 cites W2117539524 @default.
- W3144102935 cites W2128053425 @default.
- W3144102935 cites W2145494108 @default.
- W3144102935 cites W2148349024 @default.
- W3144102935 cites W2155541015 @default.
- W3144102935 cites W2159291411 @default.
- W3144102935 cites W2187089797 @default.
- W3144102935 cites W2194775991 @default.
- W3144102935 cites W2212660284 @default.
- W3144102935 cites W2279034837 @default.
- W3144102935 cites W2308529009 @default.
- W3144102935 cites W2321533354 @default.
- W3144102935 cites W2326925005 @default.
- W3144102935 cites W2412782625 @default.
- W3144102935 cites W2478454054 @default.
- W3144102935 cites W2560023338 @default.
- W3144102935 cites W2584886900 @default.
- W3144102935 cites W2593768305 @default.
- W3144102935 cites W2605488490 @default.
- W3144102935 cites W2617027347 @default.
- W3144102935 cites W2627183927 @default.
- W3144102935 cites W2766897166 @default.
- W3144102935 cites W2795155917 @default.
- W3144102935 cites W2798991696 @default.
- W3144102935 cites W2803297029 @default.
- W3144102935 cites W2842511635 @default.
- W3144102935 cites W2948127555 @default.
- W3144102935 cites W2949517790 @default.
- W3144102935 cites W2958360136 @default.
- W3144102935 cites W2962742544 @default.
- W3144102935 cites W2962793481 @default.
- W3144102935 cites W2962808524 @default.
- W3144102935 cites W2963403405 @default.
- W3144102935 cites W2963420272 @default.
- W3144102935 cites W2963446712 @default.
- W3144102935 cites W2963532621 @default.
- W3144102935 cites W2963709863 @default.
- W3144102935 cites W2963777311 @default.
- W3144102935 cites W2963784072 @default.
- W3144102935 cites W2963826681 @default.
- W3144102935 cites W2964278684 @default.
- W3144102935 cites W2970971581 @default.
- W3144102935 cites W2971155163 @default.
- W3144102935 cites W2975758481 @default.
- W3144102935 cites W2978426779 @default.
- W3144102935 cites W2981540341 @default.
- W3144102935 cites W2981720610 @default.
- W3144102935 cites W2986381065 @default.
- W3144102935 cites W2995181141 @default.
- W3144102935 cites W2996787432 @default.
- W3144102935 cites W3005680577 @default.
- W3144102935 cites W3009561768 @default.
- W3144102935 cites W3011351697 @default.
- W3144102935 cites W3013245634 @default.
- W3144102935 cites W3022061250 @default.
- W3144102935 cites W3023371261 @default.
- W3144102935 cites W3034373833 @default.
- W3144102935 cites W3034401437 @default.
- W3144102935 cites W3034781633 @default.
- W3144102935 cites W3035524453 @default.
- W3144102935 cites W3036982689 @default.
- W3144102935 cites W3039362982 @default.
- W3144102935 cites W3089515823 @default.
- W3144102935 cites W3091471537 @default.
- W3144102935 cites W3091695563 @default.
- W3144102935 cites W3095121901 @default.
- W3144102935 cites W343636949 @default.
- W3144102935 doi "https://doi.org/10.48550/arxiv.2103.16765" @default.
- W3144102935 hasPublicationYear "2021" @default.
- W3144102935 type Work @default.
- W3144102935 sameAs 3144102935 @default.
- W3144102935 citedByCount "5" @default.
- W3144102935 countsByYear W31441029352021 @default.
- W3144102935 countsByYear W31441029352022 @default.
- W3144102935 crossrefType "posted-content" @default.
- W3144102935 hasAuthorship W3144102935A5007604342 @default.
- W3144102935 hasAuthorship W3144102935A5013030532 @default.
- W3144102935 hasAuthorship W3144102935A5027804169 @default.
- W3144102935 hasAuthorship W3144102935A5029105520 @default.
- W3144102935 hasAuthorship W3144102935A5047285420 @default.
- W3144102935 hasAuthorship W3144102935A5074250521 @default.