{"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/squeezing-backbone-feature-distributions-to","title":"Squeezing Backbone Feature Distributions to the Max for Efficient Few-Shot Learning","arxiv_id":"2110.09446","date":"2021-10-18","proceeding":null,"authors":["Yuqing Hu","Vincent Gripon","Stéphane Pateux"],"abstract":"Few-shot classification is a challenging problem due to the uncertainty caused by using few labelled samples. In the past few years, many methods have been proposed with the common aim of transferring knowledge acquired on a previously solved task, what is often achieved by using a pretrained feature extractor. Following this vein, in this paper we propose a novel transfer-based method which aims at processing the feature vectors so that they become closer to Gaussian-like distributions, resulting in increased accuracy. In the case of transductive few-shot learning where unlabelled test samples are available during training, we also introduce an optimal-transport inspired algorithm to boost even further the achieved performance. Using standardized vision benchmarks, we show the ability of the proposed methodology to achieve state-of-the-art accuracy with various datasets, backbone architectures and few-shot settings.","url_abs":"https://arxiv.org/abs/2110.09446v1","url_pdf":"https://arxiv.org/pdf/2110.09446v1.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":"squeezing-backbone-feature-distributions-to","repo_url":"https://github.com/yhu01/bms","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"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"}],"methods":[{"method_slug":"test","method_name":"Test"}],"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":"PEMnE-BMS*","rank_in_archive_order":3,"of":38,"metrics":{"Accuracy":"88.44"},"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":"PEMnE-BMS*","rank_in_archive_order":5,"of":39,"metrics":{"Accuracy":"91.86"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-cub-200-5-1","task":"Few-Shot Image Classification","dataset":"CUB 200 5-way 1-shot","model":"PEMnE-BMS*","rank_in_archive_order":5,"of":36,"metrics":{"Accuracy":"94.78"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-cub-200-5","task":"Few-Shot Image Classification","dataset":"CUB 200 5-way 5-shot","model":"PEMnE-BMS*","rank_in_archive_order":4,"of":32,"metrics":{"Accuracy":"96.43"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-mini-5","task":"Few-Shot Image Classification","dataset":"Mini-ImageNet-CUB 5-way (1-shot)","model":"PEMnE-BMS*","rank_in_archive_order":2,"of":12,"metrics":{"Accuracy":"63.90"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-6","task":"Few-Shot Image Classification","dataset":"Mini-ImageNet-CUB 5-way (5-shot)","model":"PEMnE-BMS","rank_in_archive_order":2,"of":8,"metrics":{"Accuracy":"79.15"},"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":"PEMnE-BMS* (transductive)","rank_in_archive_order":7,"of":105,"metrics":{"Accuracy":"85.54"},"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":"PEMbE-NCM (inductive)","rank_in_archive_order":38,"of":105,"metrics":{"Accuracy":"68.43"},"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":"PEMnE-BMS*(transductive)","rank_in_archive_order":6,"of":95,"metrics":{"Accuracy":"91.53"},"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":"PEMbE-NCM (inductive)","rank_in_archive_order":26,"of":95,"metrics":{"Accuracy":"84.67"},"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":"PEMnE-BMS*","rank_in_archive_order":3,"of":49,"metrics":{"Accuracy":"86.07"},"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":"PEMnE-BMS*","rank_in_archive_order":4,"of":51,"metrics":{"Accuracy":"91.09"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}