{"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/190600695","title":"Continual learning with hypernetworks","arxiv_id":"1906.00695","date":"2019-06-03","proceeding":"ICLR 2020 1","authors":["Johannes von Oswald","Christian Henning","Benjamin F. Grewe","João Sacramento"],"abstract":"Artificial neural networks suffer from catastrophic forgetting when they are sequentially trained on multiple tasks. To overcome this problem, we present a novel approach based on task-conditioned hypernetworks, i.e., networks that generate the weights of a target model based on task identity. Continual learning (CL) is less difficult for this class of models thanks to a simple key feature: instead of recalling the input-output relations of all previously seen data, task-conditioned hypernetworks only require rehearsing task-specific weight realizations, which can be maintained in memory using a simple regularizer. Besides achieving state-of-the-art performance on standard CL benchmarks, additional experiments on long task sequences reveal that task-conditioned hypernetworks display a very large capacity to retain previous memories. Notably, such long memory lifetimes are achieved in a compressive regime, when the number of trainable hypernetwork weights is comparable or smaller than target network size. We provide insight into the structure of low-dimensional task embedding spaces (the input space of the hypernetwork) and show that task-conditioned hypernetworks demonstrate transfer learning. Finally, forward information transfer is further supported by empirical results on a challenging CL benchmark based on the CIFAR-10/100 image datasets.","url_abs":"https://arxiv.org/abs/1906.00695v4","url_pdf":"https://arxiv.org/pdf/1906.00695v4.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":"190600695","repo_url":"https://github.com/chrhenning/hypercl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"190600695","repo_url":"https://github.com/chrhenning/hypnettorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"190600695","repo_url":"https://github.com/gahaalt/continual-learning-overview","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"190600695","repo_url":"https://github.com/gahaalt/continual-learning-with-hypernets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"190600695","repo_url":"https://github.com/geox-lab/cmn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"190600695","repo_url":"https://github.com/gmum/hint","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"190600695","repo_url":"https://github.com/gmum/hyperinterval","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"190600695","repo_url":"https://github.com/pennfranc/hypnettorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"190600695","repo_url":"https://github.com/rvl-lab-utoronto/HyperCRL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"continual-learning","task_name":"Continual Learning"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[{"method_slug":"hypernetwork","method_name":"HyperNetwork"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/continual-learning-on-f-celeba-10-tasks","task":"Continual Learning","dataset":"F-CelebA (10 tasks)","model":"HyperNet","rank_in_archive_order":4,"of":7,"metrics":{"Acc":"0.6036"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1906.00695","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1906.00695"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gahaalt/continual-learning-overview","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gmum/hint","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gahaalt/continual-learning-with-hypernets","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/chrhenning/hypercl","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/chrhenning/hypnettorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/rvl-lab-utoronto/HyperCRL","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gmum/hyperinterval","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/geox-lab/cmn","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pennfranc/hypnettorch","reach":{"status":"unanswered"}}],"summary":{"ran_draft_wrong":1},"by_repo_kind":{"listed":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"3263fbc26d935680","entry":"work","repo":"rvl-lab-utoronto/HyperCRL","repo_kind":"listed","path":"scripts/run_pusher.py","file_url":"https://github.com/rvl-lab-utoronto/HyperCRL/blob/HEAD/scripts/run_pusher.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3263fbc26d935680"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}