{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/probabilistic-structural-latent-1","title":"Probabilistic Structural Latent Representation for Unsupervised Embedding","arxiv_id":null,"date":"2020-06-22","proceeding":null,"authors":["Mang Ye","Jianbing Shen∗"],"abstract":"Unsupervised embedding learning aims at extracting\r\nlow-dimensional visually meaningful representations from\r\nlarge-scale unlabeled images, which can then be directly\r\nused for similarity-based search. This task faces two major\r\nchallenges: 1) mining positive supervision from highly similar fine-grained classes and 2) generating to unseen testing categories. To tackle these issues, this paper proposes a\r\nprobabilistic structural latent representation (PSLR), which\r\nincorporates an adaptable softmax embedding to approximate the positive concentrated and negative instance separated properties in the graph latent space. It improves\r\nthe discriminability by enlarging the positive/negative difference without introducing any additional computational\r\ncost while maintaining high learning efficiency. To address\r\nthe limited supervision using data augmentation, a smooth\r\nvariational reconstruction loss is introduced by modeling\r\nthe intra-instance variance, which improves the robustness.\r\nExtensive experiments demonstrate the superiority of PSLR\r\nover state-of-the-art unsupervised methods on both seen\r\nand unseen categories with cosine similarity. Code is available at https://github.com/mangye16/PSLR","url_abs":"https://openaccess.thecvf.com/content_CVPR_2020/papers/Ye_Probabilistic_Structural_Latent_Representation_for_Unsupervised_Embedding_CVPR_2020_paper.pdf","url_pdf":"https://github.com/mangye16/PSLR","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"probabilistic-structural-latent-1","repo_url":"https://github.com/mangye16/PSLR","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"}],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}