Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/1904.09601 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916279684694016 |
|---|---|
| author | Tao, Chaofan Lv, Fengmao Duan, Lixin Wu, Min |
| author_facet | Tao, Chaofan Lv, Fengmao Duan, Lixin Wu, Min |
| contents | How to effectively learn from unlabeled data from the target domain is crucial for domain adaptation, as it helps reduce the large performance gap due to domain shift or distribution change. In this paper, we propose an easy-to-implement method dubbed MiniMax Entropy Networks (MMEN) based on adversarial learning. Unlike most existing approaches which employ a generator to deal with domain difference, MMEN focuses on learning the categorical information from unlabeled target samples with the help of labeled source samples. Specifically, we set an unfair multi-class classifier named categorical discriminator, which classifies source samples accurately but be confused about the categories of target samples. The generator learns a common subspace that aligns the unlabeled samples based on the target pseudo-labels. For MMEN, we also provide theoretical explanations to show that the learning of feature alignment reduces domain mismatch at the category level. Experimental results on various benchmark datasets demonstrate the effectiveness of our method over existing state-of-the-art baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_1904_09601 |
| institution | arXiv |
| publishDate | 2019 |
| record_format | arxiv |
| spellingShingle | MiniMax Entropy Network: Learning Category-Invariant Features for Domain Adaptation Tao, Chaofan Lv, Fengmao Duan, Lixin Wu, Min Computer Vision and Pattern Recognition How to effectively learn from unlabeled data from the target domain is crucial for domain adaptation, as it helps reduce the large performance gap due to domain shift or distribution change. In this paper, we propose an easy-to-implement method dubbed MiniMax Entropy Networks (MMEN) based on adversarial learning. Unlike most existing approaches which employ a generator to deal with domain difference, MMEN focuses on learning the categorical information from unlabeled target samples with the help of labeled source samples. Specifically, we set an unfair multi-class classifier named categorical discriminator, which classifies source samples accurately but be confused about the categories of target samples. The generator learns a common subspace that aligns the unlabeled samples based on the target pseudo-labels. For MMEN, we also provide theoretical explanations to show that the learning of feature alignment reduces domain mismatch at the category level. Experimental results on various benchmark datasets demonstrate the effectiveness of our method over existing state-of-the-art baselines. |
| title | MiniMax Entropy Network: Learning Category-Invariant Features for Domain Adaptation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/1904.09601 |