{"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/relay-backpropagation-for-effective-learning","title":"Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks","arxiv_id":"1512.05830","date":"2015-12-18","proceeding":null,"authors":["Li Shen","Zhouchen Lin","Qingming Huang"],"abstract":"Learning deeper convolutional neural networks becomes a tendency in recent\nyears. However, many empirical evidences suggest that performance improvement\ncannot be gained by simply stacking more layers. In this paper, we consider the\nissue from an information theoretical perspective, and propose a novel method\nRelay Backpropagation, that encourages the propagation of effective information\nthrough the network in training stage. By virtue of the method, we achieved the\nfirst place in ILSVRC 2015 Scene Classification Challenge. Extensive\nexperiments on two challenging large scale datasets demonstrate the\neffectiveness of our method is not restricted to a specific dataset or network\narchitecture. Our models will be available to the research community later.","url_abs":"http://arxiv.org/abs/1512.05830v2","url_pdf":"http://arxiv.org/pdf/1512.05830v2.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":"relay-backpropagation-for-effective-learning","repo_url":"https://github.com/craston/object_detection_cib","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"long-tail-learning","task_name":"Long-tail Learning"},{"task_slug":"scene-classification","task_name":"Scene Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/long-tail-learning-on-coco-mlt","task":"Long-tail Learning","dataset":"COCO-MLT","model":"RS(ResNet-50)","rank_in_archive_order":10,"of":13,"metrics":{"Average mAP":"46.97"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-voc-mlt","task":"Long-tail Learning","dataset":"VOC-MLT","model":"RS(ResNet-50)","rank_in_archive_order":8,"of":13,"metrics":{"Average mAP":"75.38"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1512.05830","atlas_url":"https://app.syntology.ai/?focus=1512.05830","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}