{"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/transboost-improving-the-best-imagenet","title":"TransBoost: Improving the Best ImageNet Performance using Deep Transduction","arxiv_id":"2205.13331","date":"2022-05-26","proceeding":null,"authors":["Omer Belhasin","Guy Bar-Shalom","Ran El-Yaniv"],"abstract":"This paper deals with deep transductive learning, and proposes TransBoost as a procedure for fine-tuning any deep neural model to improve its performance on any (unlabeled) test set provided at training time. TransBoost is inspired by a large margin principle and is efficient and simple to use. Our method significantly improves the ImageNet classification performance on a wide range of architectures, such as ResNets, MobileNetV3-L, EfficientNetB0, ViT-S, and ConvNext-T, leading to state-of-the-art transductive performance. Additionally we show that TransBoost is effective on a wide variety of image classification datasets. The implementation of TransBoost is provided at: https://github.com/omerb01/TransBoost .","url_abs":"https://arxiv.org/abs/2205.13331v4","url_pdf":"https://arxiv.org/pdf/2205.13331v4.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":"transboost-improving-the-best-imagenet","repo_url":"https://github.com/omerb01/transboost","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"transductive-learning","task_name":"Transductive Learning"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-cifar-10","task":"Image Classification","dataset":"CIFAR-10","model":"TransBoost-ResNet50","rank_in_archive_order":76,"of":265,"metrics":{"Percentage correct":"97.61"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-dtd","task":"Image Classification","dataset":"DTD","model":"TransBoost-ResNet50","rank_in_archive_order":9,"of":11,"metrics":{"Accuracy":"76.49"},"uses_additional_data":true},{"leaderboard":"/sota/image-classification-on-fgvc-aircraft","task":"Image Classification","dataset":"FGVC Aircraft","model":"TransBoost-ResNet50","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"83.80%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-flowers-102","task":"Image Classification","dataset":"Flowers-102","model":"TransBoost-ResNet50","rank_in_archive_order":34,"of":52,"metrics":{"Accuracy":"97.85%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-food-101-1","task":"Image Classification","dataset":"Food-101","model":"TransBoost-ResNet50","rank_in_archive_order":7,"of":11,"metrics":{"Accuracy (%)":"84.30"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ViT-S","rank_in_archive_order":402,"of":1060,"metrics":{"Number of params":"22.05M","Top 1 Accuracy":"83.67%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ConvNext-T","rank_in_archive_order":535,"of":1060,"metrics":{"Number of params":"28.59M","Top 1 Accuracy":"82.46%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-Swin-T","rank_in_archive_order":570,"of":1060,"metrics":{"Number of params":"71.71M","Top 1 Accuracy":"82.16%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ResNet50-StrikesBack","rank_in_archive_order":657,"of":1060,"metrics":{"Number of params":"25.56M","Top 1 Accuracy":"81.15%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ResNet152","rank_in_archive_order":690,"of":1060,"metrics":{"Number of params":"60.19M","Top 1 Accuracy":"80.64%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ResNet101","rank_in_archive_order":732,"of":1060,"metrics":{"Number of params":"44.55M","Top 1 Accuracy":"79.86%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ResNet50","rank_in_archive_order":788,"of":1060,"metrics":{"Top 1 Accuracy":"79.03%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-EfficientNetB0","rank_in_archive_order":820,"of":1060,"metrics":{"Number of params":"5.29M","Top 1 Accuracy":"78.60%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-MobileNetV3-L","rank_in_archive_order":896,"of":1060,"metrics":{"Number of params":"5.48M","Top 1 Accuracy":"76.81%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ResNet34","rank_in_archive_order":906,"of":1060,"metrics":{"Number of params":"21.8M","Top 1 Accuracy":"76.70%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"TransBoost-ResNet18","rank_in_archive_order":987,"of":1060,"metrics":{"Number of params":"11.69M","Top 1 Accuracy":"73.36%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-sun397","task":"Image Classification","dataset":"SUN397","model":"TransBoost-ResNet50","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"95.94%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stanford-cars","task":"Image Classification","dataset":"Stanford Cars","model":"TransBoost-ResNet50","rank_in_archive_order":13,"of":24,"metrics":{"Accuracy":"90.80%"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}