{"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/neural-optimizer-search-with-reinforcement","title":"Neural Optimizer Search with Reinforcement Learning","arxiv_id":"1709.07417","date":"2017-09-21","proceeding":null,"authors":["Irwan Bello","Barret Zoph","Vijay Vasudevan","Quoc V. Le"],"abstract":"We present an approach to automate the process of discovering optimization\nmethods, with a focus on deep learning architectures. We train a Recurrent\nNeural Network controller to generate a string in a domain specific language\nthat describes a mathematical update equation based on a list of primitive\nfunctions, such as the gradient, running average of the gradient, etc. The\ncontroller is trained with Reinforcement Learning to maximize the performance\nof a model after a few epochs. On CIFAR-10, our method discovers several update\nrules that are better than many commonly used optimizers, such as Adam,\nRMSProp, or SGD with and without Momentum on a ConvNet model. We introduce two\nnew optimizers, named PowerSign and AddSign, which we show transfer well and\nimprove training on a variety of different tasks and architectures, including\nImageNet classification and Google's neural machine translation system.","url_abs":"http://arxiv.org/abs/1709.07417v2","url_pdf":"http://arxiv.org/pdf/1709.07417v2.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":"neural-optimizer-search-with-reinforcement","repo_url":"https://github.com/calclavia/NOS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"neural-optimizer-search-with-reinforcement","repo_url":"https://github.com/carpedm20/ENAS-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"sgd","method_name":"SGD"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1709.07417","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}