{"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/task2vec-task-embedding-for-meta-learning","title":"Task2Vec: Task Embedding for Meta-Learning","arxiv_id":"1902.03545","date":"2019-02-10","proceeding":"ICCV 2019 10","authors":["Alessandro Achille","Michael Lam","Rahul Tewari","Avinash Ravichandran","Subhransu Maji","Charless Fowlkes","Stefano Soatto","Pietro Perona"],"abstract":"We introduce a method to provide vectorial representations of visual\nclassification tasks which can be used to reason about the nature of those\ntasks and their relations. Given a dataset with ground-truth labels and a loss\nfunction defined over those labels, we process images through a \"probe network\"\nand compute an embedding based on estimates of the Fisher information matrix\nassociated with the probe network parameters. This provides a fixed-dimensional\nembedding of the task that is independent of details such as the number of\nclasses and does not require any understanding of the class label semantics. We\ndemonstrate that this embedding is capable of predicting task similarities that\nmatch our intuition about semantic and taxonomic relations between different\nvisual tasks (e.g., tasks based on classifying different types of plants are\nsimilar) We also demonstrate the practical value of this framework for the\nmeta-task of selecting a pre-trained feature extractor for a new task. We\npresent a simple meta-learning framework for learning a metric on embeddings\nthat is capable of predicting which feature extractors will perform well.\nSelecting a feature extractor with task embedding obtains a performance close\nto the best available feature extractor, while costing substantially less than\nexhaustively training and evaluating on all available feature extractors.","url_abs":"http://arxiv.org/abs/1902.03545v1","url_pdf":"http://arxiv.org/pdf/1902.03545v1.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":"task2vec-task-embedding-for-meta-learning","repo_url":"https://github.com/awslabs/aws-cv-task2vec","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"meta-learning","task_name":"Meta-Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1902.03545","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1902.03545"}},"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/awslabs/aws-cv-task2vec","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":16},"by_repo_kind":{"listed":{"samples":16,"ran":0,"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":"9306127e388d80b3","entry":"average_top_k_tax_distance","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"plot_distance_cub_inat.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/plot_distance_cub_inat.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9306127e388d80b3"}},{"code_sha256_prefix":"944bd91f1edbc80d","entry":"get_compression_loss","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"variational.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/variational.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"944bd91f1edbc80d"}},{"code_sha256_prefix":"0c35c102a078d24d","entry":"get_device","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0c35c102a078d24d"}},{"code_sha256_prefix":"525c6605bb5ab89a","entry":"get_error","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"525c6605bb5ab89a"}},{"code_sha256_prefix":"052a7c3d285c564d","entry":"get_image","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"plot_distance_cub_inat.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/plot_distance_cub_inat.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"052a7c3d285c564d"}},{"code_sha256_prefix":"8846b271d3785538","entry":"get_model","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"models.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/models.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8846b271d3785538"}},{"code_sha256_prefix":"92003a5021d20c3d","entry":"get_trivial_embedding_from","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"task_similarity.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/task_similarity.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"92003a5021d20c3d"}},{"code_sha256_prefix":"dbe657c4babc8ae6","entry":"get_variational_vars","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"variational.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/variational.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"dbe657c4babc8ae6"}},{"code_sha256_prefix":"7cc3548449813ac3","entry":"inat2018","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"datasets.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/datasets.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7cc3548449813ac3"}},{"code_sha256_prefix":"ebe075162f279d94","entry":"invert_icon","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"plot_distance_cub_inat.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/plot_distance_cub_inat.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ebe075162f279d94"}},{"code_sha256_prefix":"c2823ae7a8245c3c","entry":"is_excluded","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"task_similarity.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/task_similarity.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c2823ae7a8245c3c"}},{"code_sha256_prefix":"dd3befed50865cc3","entry":"is_multi_label","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"dataset/dataset.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/dataset/dataset.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"dd3befed50865cc3"}},{"code_sha256_prefix":"7a95ae77a9468ef9","entry":"load_embedding","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"task_similarity.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/task_similarity.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7a95ae77a9468ef9"}},{"code_sha256_prefix":"97198064440d5f0f","entry":"load_tasks_map","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"datasets.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/datasets.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"97198064440d5f0f"}},{"code_sha256_prefix":"ca9c797073c0672e","entry":"set_metadata","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"datasets.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/datasets.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ca9c797073c0672e"}},{"code_sha256_prefix":"783f27dc9447605e","entry":"variational_forward","repo":"awslabs/aws-cv-task2vec","repo_kind":"listed","path":"variational.py","file_url":"https://github.com/awslabs/aws-cv-task2vec/blob/HEAD/variational.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"783f27dc9447605e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}