{"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/prototype-completion-with-primitive-knowledge","title":"Prototype Completion with Primitive Knowledge for Few-Shot Learning","arxiv_id":"2009.04960","date":"2020-09-10","proceeding":"CVPR 2021 1","authors":["Baoquan Zhang","Xutao Li","Yunming Ye","Zhichao Huang","Lisai Zhang"],"abstract":"Few-shot learning is a challenging task, which aims to learn a classifier for novel classes with few examples. Pre-training based meta-learning methods effectively tackle the problem by pre-training a feature extractor and then fine-tuning it through the nearest centroid based meta-learning. However, results show that the fine-tuning step makes very marginal improvements. In this paper, 1) we figure out the key reason, i.e., in the pre-trained feature space, the base classes already form compact clusters while novel classes spread as groups with large variances, which implies that fine-tuning the feature extractor is less meaningful; 2) instead of fine-tuning the feature extractor, we focus on estimating more representative prototypes during meta-learning. Consequently, we propose a novel prototype completion based meta-learning framework. This framework first introduces primitive knowledge (i.e., class-level part or attribute annotations) and extracts representative attribute features as priors. Then, we design a prototype completion network to learn to complete prototypes with these priors. To avoid the prototype completion error caused by primitive knowledge noises or class differences, we further develop a Gaussian based prototype fusion strategy that combines the mean-based and completed prototypes by exploiting the unlabeled samples. Extensive experiments show that our method: (i) can obtain more accurate prototypes; (ii) outperforms state-of-the-art techniques by 2% - 9% in terms of classification accuracy. Our code is available online.","url_abs":"https://arxiv.org/abs/2009.04960v6","url_pdf":"https://arxiv.org/pdf/2009.04960v6.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":"prototype-completion-with-primitive-knowledge","repo_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"meta-learning","task_name":"Meta-Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2009.04960","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2009.04960"}},"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/zhangbq-research/Prototype_Completion_for_FSL","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":3,"ran_draft_wrong":1,"unverified":9},"by_repo_kind":{"official":{"samples":13,"ran":4,"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":"de92832114f71d8c","entry":"conv1x1","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"models/resnet12_2.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/models/resnet12_2.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"de92832114f71d8c"}},{"code_sha256_prefix":"e5b7c79b0b62af96","entry":"conv3x3","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"models/resnet12_2.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/models/resnet12_2.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e5b7c79b0b62af96"}},{"code_sha256_prefix":"f36ac42575cd46ee","entry":"norm_layer","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"models/resnet12_2.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/models/resnet12_2.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f36ac42575cd46ee"}},{"code_sha256_prefix":"7c64c35be7f02395","entry":"one_hot","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"models/classification_heads.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/models/classification_heads.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7c64c35be7f02395"}},{"code_sha256_prefix":"068a0a6a30e1e25c","entry":"CosineNetHead","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"models/classification_heads.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/models/classification_heads.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":"068a0a6a30e1e25c"}},{"code_sha256_prefix":"4ec52ee88c7bc858","entry":"CosineNetHead","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"models/classification_heads_orgin.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/models/classification_heads_orgin.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":"4ec52ee88c7bc858"}},{"code_sha256_prefix":"f7312cd2956a3b29","entry":"FuseCosineNetHead","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"models/classification_heads.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/models/classification_heads.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":"f7312cd2956a3b29"}},{"code_sha256_prefix":"161f62eb1a752b4c","entry":"constructedges","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"prior/make_miniimagenet_primitive_knowledge.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/prior/make_miniimagenet_primitive_knowledge.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":"161f62eb1a752b4c"}},{"code_sha256_prefix":"fb61241b42c06d73","entry":"count_accuracy","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"utils.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/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":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fb61241b42c06d73"}},{"code_sha256_prefix":"0c7425262620b51b","entry":"getnode","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"prior/make_miniimagenet_primitive_knowledge.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/prior/make_miniimagenet_primitive_knowledge.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":"0c7425262620b51b"}},{"code_sha256_prefix":"212f1ff29913b717","entry":"getwnid","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"prior/make_miniimagenet_primitive_knowledge.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/prior/make_miniimagenet_primitive_knowledge.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":"212f1ff29913b717"}},{"code_sha256_prefix":"78ac9e19f328f531","entry":"l2_loss","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"utils.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/utils.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":"78ac9e19f328f531"}},{"code_sha256_prefix":"a2b31023f6121fdf","entry":"pick_vectors","repo":"zhangbq-research/Prototype_Completion_for_FSL","repo_kind":"official","path":"utils.py","file_url":"https://github.com/zhangbq-research/Prototype_Completion_for_FSL/blob/HEAD/utils.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":"a2b31023f6121fdf"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}