{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/rapid-learning-or-feature-reuse-towards","title":"Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML","arxiv_id":"1909.09157","date":"2019-09-19","proceeding":"ICLR 2020 1","authors":["Aniruddh Raghu","Maithra Raghu","Samy Bengio","Oriol Vinyals"],"abstract":"An important research direction in machine learning has centered around developing meta-learning algorithms to tackle few-shot learning. An especially successful algorithm has been Model Agnostic Meta-Learning (MAML), a method that consists of two optimization loops, with the outer loop finding a meta-initialization, from which the inner loop can efficiently learn new tasks. Despite MAML's popularity, a fundamental open question remains -- is the effectiveness of MAML due to the meta-initialization being primed for rapid learning (large, efficient changes in the representations) or due to feature reuse, with the meta initialization already containing high quality features? We investigate this question, via ablation studies and analysis of the latent representations, finding that feature reuse is the dominant factor. This leads to the ANIL (Almost No Inner Loop) algorithm, a simplification of MAML where we remove the inner loop for all but the (task-specific) head of a MAML-trained network. ANIL matches MAML's performance on benchmark few-shot image classification and RL and offers computational improvements over MAML. We further study the precise contributions of the head and body of the network, showing that performance on the test tasks is entirely determined by the quality of the learned features, and we can remove even the head of the network (the NIL algorithm). We conclude with a discussion of the rapid learning vs feature reuse question for meta-learning algorithms more broadly.","url_abs":"https://arxiv.org/abs/1909.09157v2","url_pdf":"https://arxiv.org/pdf/1909.09157v2.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":"rapid-learning-or-feature-reuse-towards","repo_url":"https://github.com/fmu2/PyTorch-MAML","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rapid-learning-or-feature-reuse-towards","repo_url":"https://github.com/learnables/learn2learn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"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":"image-classification","task_name":"Image Classification"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"open-question","task_name":"Open-Ended Question Answering"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"maml","method_name":"MAML"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1909.09157","atlas_url":"https://app.syntology.ai/?focus=1909.09157","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"1909.09157"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"1909.09157\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"1909.09157"},"call":"get_concepts_for_paper(arxiv_id=\"1909.09157\")"}],"by_repo":[{"repo":"fmu2/PyTorch-MAML","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0,"harvested_for_other_papers":1,"harvested_for":["1703.03400"],"harvested_for_calls":[{"arxiv_id":"1703.03400","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"1703.03400"},"call":"get_harvested_code_for_paper(arxiv_id=\"1703.03400\")"}]},{"repo":"learnables/learn2learn","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0,"harvested_for_other_papers":4,"harvested_for":["1703.03400","1703.05175","1803.02999","2008.12284"],"harvested_for_calls":[{"arxiv_id":"1703.03400","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"1703.03400"},"call":"get_harvested_code_for_paper(arxiv_id=\"1703.03400\")"},{"arxiv_id":"1703.05175","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"1703.05175"},"call":"get_harvested_code_for_paper(arxiv_id=\"1703.05175\")"},{"arxiv_id":"1803.02999","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"1803.02999"},"call":"get_harvested_code_for_paper(arxiv_id=\"1803.02999\")"},{"arxiv_id":"2008.12284","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2008.12284"},"call":"get_harvested_code_for_paper(arxiv_id=\"2008.12284\")"}]}],"by_repo_harvested_for_is":"harvested_for: the arXiv ids (as Syntology's graph holds them) of every paper whose samples Syntology's graph links to this repository, in arXiv-id order, complete; the page names the first 3 and says how many more. A link only: it does not say the papers are the same work, that one copied the other, or that the code implements this paper's method; harvested_for_calls gives the get_harvested_code_for_paper call for each arXiv-shaped id among them","by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"b99e9de00047bca14033663b88117a0c8dfb8d32ba5c3f35ce6b46ee39e52937","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}