{"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/overcoming-catastrophic-forgetting-by","title":"Overcoming Catastrophic Forgetting by Incremental Moment Matching","arxiv_id":"1703.08475","date":"2017-03-24","proceeding":"NeurIPS 2017 12","authors":["Sang-Woo Lee","Jin-Hwa Kim","Jaehyun Jun","Jung-Woo Ha","Byoung-Tak Zhang"],"abstract":"Catastrophic forgetting is a problem of neural networks that loses the\ninformation of the first task after training the second task. Here, we propose\na method, i.e. incremental moment matching (IMM), to resolve this problem. IMM\nincrementally matches the moment of the posterior distribution of the neural\nnetwork which is trained on the first and the second task, respectively. To\nmake the search space of posterior parameter smooth, the IMM procedure is\ncomplemented by various transfer learning techniques including weight transfer,\nL2-norm of the old and the new parameter, and a variant of dropout with the old\nparameter. We analyze our approach on a variety of datasets including the\nMNIST, CIFAR-10, Caltech-UCSD-Birds, and Lifelog datasets. The experimental\nresults show that IMM achieves state-of-the-art performance by balancing the\ninformation between an old and a new network.","url_abs":"http://arxiv.org/abs/1703.08475v3","url_pdf":"http://arxiv.org/pdf/1703.08475v3.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":"overcoming-catastrophic-forgetting-by","repo_url":"https://github.com/btjhjeon/IMM_tensorflow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[{"method_slug":"dropout","method_name":"Dropout"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.08475","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}