{"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/training-deep-neural-networks-via-direct-loss","title":"Training Deep Neural Networks via Direct Loss Minimization","arxiv_id":"1511.06411","date":"2015-11-19","proceeding":null,"authors":["Yang Song","Alexander G. Schwing","Richard S. Zemel","Raquel Urtasun"],"abstract":"Supervised training of deep neural nets typically relies on minimizing\ncross-entropy. However, in many domains, we are interested in performing well\non metrics specific to the application. In this paper we propose a direct loss\nminimization approach to train deep neural networks, which provably minimizes\nthe application-specific loss function. This is often non-trivial, since these\nfunctions are neither smooth nor decomposable and thus are not amenable to\noptimization with standard gradient-based methods. We demonstrate the\neffectiveness of our approach in the context of maximizing average precision\nfor ranking problems. Towards this goal, we develop a novel dynamic programming\nalgorithm that can efficiently compute the weight updates. Our approach proves\nsuperior to a variety of baselines in the context of action classification and\nobject detection, especially in the presence of label noise.","url_abs":"http://arxiv.org/abs/1511.06411v2","url_pdf":"http://arxiv.org/pdf/1511.06411v2.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":"training-deep-neural-networks-via-direct-loss","repo_url":"https://github.com/yang-song/APDLM","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"action-classification","task_name":"Action Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1511.06411","atlas_url":"https://app.syntology.ai/?focus=1511.06411","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}