Summary of Adaembed: Semi-supervised Domain Adaptation in the Embedding Space, by Ali Mottaghi et al.
AdaEmbed: Semi-supervised Domain Adaptation in the Embedding Space
by Ali Mottaghi, Mohammad Abdullah Jamal, Serena Yeung, Omid Mohareri
First submitted to arxiv on: 23 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed methodology, AdaEmbed, tackles the challenge of semi-supervised domain adaptation (SSDA) in computer vision by leveraging unlabeled data to transfer knowledge from a labeled source domain to an unlabeled target domain. By learning a shared embedding space and generating accurate pseudo-labels, AdaEmbed outperforms baselines on benchmark datasets like DomainNet, Office-Home, and VisDA-C, setting a new state of the art for SSDA. This approach is particularly useful in real-world scenarios where labeled data is scarce. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary AdaEmbed helps computers learn from one domain and apply it to another without much labeled data. It does this by creating a shared space that connects both domains and gives a good guess at what labels should be. This works really well on big datasets like DomainNet, Office-Home, and VisDA-C, and even beats the previous best methods. This is important because in real life, we often don’t have enough labeled data to train a model. |
Keywords
» Artificial intelligence » Domain adaptation » Embedding space » Semi supervised
