Conditional Adversarial Domain Adaptation. Long, M., CAO, Z., Wang, J., & Jordan, M. I In Advances in Neural Information Processing Systems, volume 31, 2018. Curran Associates, Inc.. ![Conditional Adversarial Domain Adaptation [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex Adversarial learning has been embedded into deep networks to learn disentangled and transferable representations for domain adaptation. Existing adversarial domain adaptation methods may struggle to align different domains of multimodal distributions that are native in classification problems. In this paper, we present conditional adversarial domain adaptation, a principled framework that conditions the adversarial adaptation models on discriminative information conveyed in the classifier predictions. Conditional domain adversarial networks (CDANs) are designed with two novel conditioning strategies: multilinear conditioning that captures the cross-covariance between feature representations and classifier predictions to improve the discriminability, and entropy conditioning that controls the uncertainty of classifier predictions to guarantee the transferability. Experiments testify that the proposed approach exceeds the state-of-the-art results on five benchmark datasets.
@inproceedings{long_conditional_2018,
title = {Conditional {Adversarial} {Domain} {Adaptation}},
volume = {31},
url = {https://papers.nips.cc/paper_files/paper/2018/hash/ab88b15733f543179858600245108dd8-Abstract.html},
abstract = {Adversarial learning has been embedded into deep networks to learn disentangled and transferable representations for domain adaptation. Existing adversarial domain adaptation methods may struggle to align different domains of multimodal distributions that are native in classification problems. In this paper, we present conditional adversarial domain adaptation, a principled framework that conditions the adversarial adaptation models on discriminative information conveyed in the classifier predictions. Conditional domain adversarial networks (CDANs) are designed with two novel conditioning strategies: multilinear conditioning that captures the cross-covariance between feature representations and classifier predictions to improve the discriminability, and entropy conditioning that controls the uncertainty of classifier predictions to guarantee the transferability. Experiments testify that the proposed approach exceeds the state-of-the-art results on five benchmark datasets.},
urldate = {2024-05-31},
booktitle = {Advances in {Neural} {Information} {Processing} {Systems}},
publisher = {Curran Associates, Inc.},
author = {Long, Mingsheng and CAO, ZHANGJIE and Wang, Jianmin and Jordan, Michael I},
year = {2018},
keywords = {⛔ No DOI found},
}
Downloads: 0
{"_id":"xFY7QjTTZa4GdjDm3","bibbaseid":"long-cao-wang-jordan-conditionaladversarialdomainadaptation-2018","author_short":["Long, M.","CAO, Z.","Wang, J.","Jordan, M. I"],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Conditional Adversarial Domain Adaptation","volume":"31","url":"https://papers.nips.cc/paper_files/paper/2018/hash/ab88b15733f543179858600245108dd8-Abstract.html","abstract":"Adversarial learning has been embedded into deep networks to learn disentangled and transferable representations for domain adaptation. Existing adversarial domain adaptation methods may struggle to align different domains of multimodal distributions that are native in classification problems. In this paper, we present conditional adversarial domain adaptation, a principled framework that conditions the adversarial adaptation models on discriminative information conveyed in the classifier predictions. Conditional domain adversarial networks (CDANs) are designed with two novel conditioning strategies: multilinear conditioning that captures the cross-covariance between feature representations and classifier predictions to improve the discriminability, and entropy conditioning that controls the uncertainty of classifier predictions to guarantee the transferability. Experiments testify that the proposed approach exceeds the state-of-the-art results on five benchmark datasets.","urldate":"2024-05-31","booktitle":"Advances in Neural Information Processing Systems","publisher":"Curran Associates, Inc.","author":[{"propositions":[],"lastnames":["Long"],"firstnames":["Mingsheng"],"suffixes":[]},{"propositions":[],"lastnames":["CAO"],"firstnames":["ZHANGJIE"],"suffixes":[]},{"propositions":[],"lastnames":["Wang"],"firstnames":["Jianmin"],"suffixes":[]},{"propositions":[],"lastnames":["Jordan"],"firstnames":["Michael","I"],"suffixes":[]}],"year":"2018","keywords":"⛔ No DOI found","bibtex":"@inproceedings{long_conditional_2018,\n\ttitle = {Conditional {Adversarial} {Domain} {Adaptation}},\n\tvolume = {31},\n\turl = {https://papers.nips.cc/paper_files/paper/2018/hash/ab88b15733f543179858600245108dd8-Abstract.html},\n\tabstract = {Adversarial learning has been embedded into deep networks to learn disentangled and transferable representations for domain adaptation. Existing adversarial domain adaptation methods may struggle to align different domains of multimodal distributions that are native in classification problems. In this paper, we present conditional adversarial domain adaptation, a principled framework that conditions the adversarial adaptation models on discriminative information conveyed in the classifier predictions. Conditional domain adversarial networks (CDANs) are designed with two novel conditioning strategies: multilinear conditioning that captures the cross-covariance between feature representations and classifier predictions to improve the discriminability, and entropy conditioning that controls the uncertainty of classifier predictions to guarantee the transferability. Experiments testify that the proposed approach exceeds the state-of-the-art results on five benchmark datasets.},\n\turldate = {2024-05-31},\n\tbooktitle = {Advances in {Neural} {Information} {Processing} {Systems}},\n\tpublisher = {Curran Associates, Inc.},\n\tauthor = {Long, Mingsheng and CAO, ZHANGJIE and Wang, Jianmin and Jordan, Michael I},\n\tyear = {2018},\n\tkeywords = {⛔ No DOI found},\n}\n\n\n\n","author_short":["Long, M.","CAO, Z.","Wang, J.","Jordan, M. I"],"key":"long_conditional_2018","id":"long_conditional_2018","bibbaseid":"long-cao-wang-jordan-conditionaladversarialdomainadaptation-2018","role":"author","urls":{"Paper":"https://papers.nips.cc/paper_files/paper/2018/hash/ab88b15733f543179858600245108dd8-Abstract.html"},"keyword":["⛔ No DOI found"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/fsimonetta","dataSources":["pzyFFGWvxG2bs63zP"],"keywords":["⛔ no doi found"],"search_terms":["conditional","adversarial","domain","adaptation","long","cao","wang","jordan"],"title":"Conditional Adversarial Domain Adaptation","year":2018}