{"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/bridging-multi-task-learning-and-meta","title":"Bridging Multi-Task Learning and Meta-Learning: Towards Efficient Training and Effective Adaptation","arxiv_id":"2106.09017","date":"2021-06-16","proceeding":null,"authors":["Haoxiang Wang","Han Zhao","Bo Li"],"abstract":"Multi-task learning (MTL) aims to improve the generalization of several related tasks by learning them jointly. As a comparison, in addition to the joint training scheme, modern meta-learning allows unseen tasks with limited labels during the test phase, in the hope of fast adaptation over them. Despite the subtle difference between MTL and meta-learning in the problem formulation, both learning paradigms share the same insight that the shared structure between existing training tasks could lead to better generalization and adaptation. In this paper, we take one important step further to understand the close connection between these two learning paradigms, through both theoretical analysis and empirical investigation. Theoretically, we first demonstrate that MTL shares the same optimization formulation with a class of gradient-based meta-learning (GBML) algorithms. We then prove that for over-parameterized neural networks with sufficient depth, the learned predictive functions of MTL and GBML are close. In particular, this result implies that the predictions given by these two models are similar over the same unseen task. Empirically, we corroborate our theoretical findings by showing that, with proper implementation, MTL is competitive against state-of-the-art GBML algorithms on a set of few-shot image classification benchmarks. Since existing GBML algorithms often involve costly second-order bi-level optimization, our first-order MTL method is an order of magnitude faster on large-scale datasets such as mini-ImageNet. We believe this work could help bridge the gap between these two learning paradigms, and provide a computationally efficient alternative to GBML that also supports fast task adaptation.","url_abs":"https://arxiv.org/abs/2106.09017v1","url_pdf":"https://arxiv.org/pdf/2106.09017v1.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":"bridging-multi-task-learning-and-meta","repo_url":"https://github.com/AI-secure/multi-task-learning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"few-shot-image-classification","task_name":"Few-Shot Image Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/few-shot-image-classification-on-cifar-fs-5","task":"Few-Shot Image Classification","dataset":"CIFAR-FS 5-way (1-shot)","model":"Multi-Task Learning","rank_in_archive_order":34,"of":38,"metrics":{"Accuracy":"69.5"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-cifar-fs-5-1","task":"Few-Shot Image Classification","dataset":"CIFAR-FS 5-way (5-shot)","model":"Multi-Task Learning","rank_in_archive_order":34,"of":39,"metrics":{"Accuracy":"84.1"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-fc100-5-way","task":"Few-Shot Image Classification","dataset":"FC100 5-way (1-shot)","model":"Multi-Task Learning","rank_in_archive_order":18,"of":22,"metrics":{"Accuracy":"42.4"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-fc100-5-way-1","task":"Few-Shot Image Classification","dataset":"FC100 5-way (5-shot)","model":"Multi-Task Learning","rank_in_archive_order":19,"of":22,"metrics":{"Accuracy":"57.7"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-2","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 5-way (1-shot)","model":"Multi-Task Learning","rank_in_archive_order":75,"of":105,"metrics":{"Accuracy":"59.84"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-3","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 5-way (5-shot)","model":"Multi-Task Learning","rank_in_archive_order":63,"of":95,"metrics":{"Accuracy":"77.72"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 5-way (1-shot)","model":"Multi-Task Learning","rank_in_archive_order":41,"of":49,"metrics":{"Accuracy":"67.11"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-1","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 5-way (5-shot)","model":"Multi-Task Learning","rank_in_archive_order":35,"of":51,"metrics":{"Accuracy":"83.69"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2106.09017","atlas_url":"https://app.syntology.ai/?focus=2106.09017","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.09017"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/AI-secure/multi-task-learning","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"unverified":9},"by_repo_kind":{"official":{"samples":10,"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":0,"samples":[{"code_sha256_prefix":"fac5364e2f53c6db","entry":"conv3x3","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"models/resnet.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/models/resnet.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fac5364e2f53c6db"}},{"code_sha256_prefix":"6c800ceb85f1eba4","entry":"create_model","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"models/util.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/models/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6c800ceb85f1eba4"}},{"code_sha256_prefix":"208f88f5e39a7ae8","entry":"get_query_support","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"lightning_episodic_module.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/lightning_episodic_module.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"208f88f5e39a7ae8"}},{"code_sha256_prefix":"b37217a4a4877177","entry":"get_teacher_name","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"models/util.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/models/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b37217a4a4877177"}},{"code_sha256_prefix":"b5562842402a79ff","entry":"index_classes","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"datasets/mini_imagenet.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/datasets/mini_imagenet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b5562842402a79ff"}},{"code_sha256_prefix":"b33624fa65d2bfb9","entry":"normalize","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"lightning_episodic_module.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/lightning_episodic_module.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b33624fa65d2bfb9"}},{"code_sha256_prefix":"df4df7ccf0516f04","entry":"resnet12","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"models/resnet.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/models/resnet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"df4df7ccf0516f04"}},{"code_sha256_prefix":"4f480d1a742fbccf","entry":"resnet18","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"models/resnet.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/models/resnet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4f480d1a742fbccf"}},{"code_sha256_prefix":"afa5a2b72957f69e","entry":"resnet50","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"models/resnet_new.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/models/resnet_new.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"afa5a2b72957f69e"}},{"code_sha256_prefix":"01ce97bf428c6241","entry":"wrn_28_10","repo":"AI-secure/multi-task-learning","repo_kind":"official","path":"models/wresnet.py","file_url":"https://github.com/AI-secure/multi-task-learning/blob/HEAD/models/wresnet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"01ce97bf428c6241"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}