{"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/learning-to-propagate-labels-transductive","title":"Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learning","arxiv_id":"1805.10002","date":"2018-05-25","proceeding":"ICLR 2019 5","authors":["Yanbin Liu","Juho Lee","Minseop Park","Saehoon Kim","Eunho Yang","Sung Ju Hwang","Yi Yang"],"abstract":"The goal of few-shot learning is to learn a classifier that generalizes well\neven when trained with a limited number of training instances per class. The\nrecently introduced meta-learning approaches tackle this problem by learning a\ngeneric classifier across a large number of multiclass classification tasks and\ngeneralizing the model to a new task. Yet, even with such meta-learning, the\nlow-data problem in the novel classification task still remains. In this paper,\nwe propose Transductive Propagation Network (TPN), a novel meta-learning\nframework for transductive inference that classifies the entire test set at\nonce to alleviate the low-data problem. Specifically, we propose to learn to\npropagate labels from labeled instances to unlabeled test instances, by\nlearning a graph construction module that exploits the manifold structure in\nthe data. TPN jointly learns both the parameters of feature embedding and the\ngraph construction in an end-to-end manner. We validate TPN on multiple\nbenchmark datasets, on which it largely outperforms existing few-shot learning\napproaches and achieves the state-of-the-art results.","url_abs":"http://arxiv.org/abs/1805.10002v5","url_pdf":"http://arxiv.org/pdf/1805.10002v5.pdf","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":"learning-to-propagate-labels-transductive","repo_url":"https://github.com/csyanbin/TPN-pytorch","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"learning-to-propagate-labels-transductive","repo_url":"https://github.com/csyanbin/TPN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"few-shot-image-classification","task_name":"Few-Shot Image Classification"},{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"graph-construction","task_name":"graph construction"}],"methods":[{"method_slug":"transductive-inference","method_name":"Transductive Inference"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/few-shot-image-classification-on-mini-12","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (1-shot)","model":"TPN (Higher Shot)","rank_in_archive_order":5,"of":14,"metrics":{"Accuracy":"38.4"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-12","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (1-shot)","model":"Label Propagation","rank_in_archive_order":7,"of":14,"metrics":{"Accuracy":"35.2"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-13","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (5-shot)","model":"TPN (Higher Shot)","rank_in_archive_order":6,"of":14,"metrics":{"Accuracy":"52.8"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-13","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (5-shot)","model":"Label Propagation","rank_in_archive_order":7,"of":14,"metrics":{"Accuracy":"51.2"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-2","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (1-shot)","model":"TPN (Higher Shot)","rank_in_archive_order":5,"of":13,"metrics":{"Accuracy":"44.8"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-2","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (1-shot)","model":"Label Propagation","rank_in_archive_order":6,"of":13,"metrics":{"Accuracy":"39.4"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-3","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (5-shot)","model":"TPN (Higher Shot)","rank_in_archive_order":5,"of":13,"metrics":{"Accuracy":"59.4"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-3","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (5-shot)","model":"Label Propagation","rank_in_archive_order":8,"of":13,"metrics":{"Accuracy":"57.9"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1805.10002","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}