{"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/hybridnet-classification-and-reconstruction","title":"HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning","arxiv_id":"1807.11407","date":"2018-07-30","proceeding":"ECCV 2018 9","authors":["Thomas Robert","Nicolas Thome","Matthieu Cord"],"abstract":"In this paper, we introduce a new model for leveraging unlabeled data to\nimprove generalization performances of image classifiers: a two-branch\nencoder-decoder architecture called HybridNet. The first branch receives\nsupervision signal and is dedicated to the extraction of invariant\nclass-related representations. The second branch is fully unsupervised and\ndedicated to model information discarded by the first branch to reconstruct\ninput data. To further support the expected behavior of our model, we propose\nan original training objective. It favors stability in the discriminative\nbranch and complementarity between the learned representations in the two\nbranches. HybridNet is able to outperform state-of-the-art results on CIFAR-10,\nSVHN and STL-10 in various semi-supervised settings. In addition,\nvisualizations and ablation studies validate our contributions and the behavior\nof the model on both CIFAR-10 and STL-10 datasets.","url_abs":"http://arxiv.org/abs/1807.11407v1","url_pdf":"http://arxiv.org/pdf/1807.11407v1.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":[],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"HybridNet","rank_in_archive_order":52,"of":117,"metrics":{"Percentage correct":"84.10"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"ResNet baseline","rank_in_archive_order":58,"of":117,"metrics":{"Percentage correct":"82.00"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"SWWAE","rank_in_archive_order":77,"of":117,"metrics":{"Percentage correct":"74.33"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1807.11407","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}