{"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/device-placement-optimization-with","title":"Device Placement Optimization with Reinforcement Learning","arxiv_id":"1706.04972","date":"2017-06-13","proceeding":"ICML 2017 8","authors":["Azalia Mirhoseini","Hieu Pham","Quoc V. Le","Benoit Steiner","Rasmus Larsen","Yuefeng Zhou","Naveen Kumar","Mohammad Norouzi","Samy Bengio","Jeff Dean"],"abstract":"The past few years have witnessed a growth in size and computational\nrequirements for training and inference with neural networks. Currently, a\ncommon approach to address these requirements is to use a heterogeneous\ndistributed environment with a mixture of hardware devices such as CPUs and\nGPUs. Importantly, the decision of placing parts of the neural models on\ndevices is often made by human experts based on simple heuristics and\nintuitions. In this paper, we propose a method which learns to optimize device\nplacement for TensorFlow computational graphs. Key to our method is the use of\na sequence-to-sequence model to predict which subsets of operations in a\nTensorFlow graph should run on which of the available devices. The execution\ntime of the predicted placements is then used as the reward signal to optimize\nthe parameters of the sequence-to-sequence model. Our main result is that on\nInception-V3 for ImageNet classification, and on RNN LSTM, for language\nmodeling and neural machine translation, our model finds non-trivial device\nplacements that outperform hand-crafted heuristics and traditional algorithmic\nmethods.","url_abs":"http://arxiv.org/abs/1706.04972v2","url_pdf":"http://arxiv.org/pdf/1706.04972v2.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":"device-placement-optimization-with","repo_url":"https://github.com/indrajeet95/Device-Placement-Optimization-with-Reinforcement-Learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"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":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.04972","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}