{"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/gradient-based-meta-learning-with-learned","title":"Gradient-Based Meta-Learning with Learned Layerwise Metric and Subspace","arxiv_id":"1801.05558","date":"2018-01-17","proceeding":"ICML 2018 7","authors":["Yoonho Lee","Seungjin Choi"],"abstract":"Gradient-based meta-learning methods leverage gradient descent to learn the\ncommonalities among various tasks. While previous such methods have been\nsuccessful in meta-learning tasks, they resort to simple gradient descent\nduring meta-testing. Our primary contribution is the {\\em MT-net}, which\nenables the meta-learner to learn on each layer's activation space a subspace\nthat the task-specific learner performs gradient descent on. Additionally, a\ntask-specific learner of an {\\em MT-net} performs gradient descent with respect\nto a meta-learned distance metric, which warps the activation space to be more\nsensitive to task identity. We demonstrate that the dimension of this learned\nsubspace reflects the complexity of the task-specific learner's adaptation\ntask, and also that our model is less sensitive to the choice of initial\nlearning rates than previous gradient-based meta-learning methods. Our method\nachieves state-of-the-art or comparable performance on few-shot classification\nand regression tasks.","url_abs":"http://arxiv.org/abs/1801.05558v3","url_pdf":"http://arxiv.org/pdf/1801.05558v3.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":"gradient-based-meta-learning-with-learned","repo_url":"https://github.com/yoonholee/MT-net","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"few-shot-image-classification","task_name":"Few-Shot Image Classification"},{"task_slug":"meta-learning","task_name":"Meta-Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/few-shot-image-classification-on-mini-2","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 5-way (1-shot)","model":"MT-Net","rank_in_archive_order":93,"of":105,"metrics":{"Accuracy":"51.7"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-omniglot-1-1","task":"Few-Shot Image Classification","dataset":"OMNIGLOT - 1-Shot, 20-way","model":"MT-net","rank_in_archive_order":9,"of":20,"metrics":{"Accuracy":"96.2%"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-omniglot-1-2","task":"Few-Shot Image Classification","dataset":"OMNIGLOT - 1-Shot, 5-way","model":"MT-net","rank_in_archive_order":5,"of":17,"metrics":{"Accuracy":"99.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1801.05558","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}