{"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/darla-improving-zero-shot-transfer-in","title":"DARLA: Improving Zero-Shot Transfer in Reinforcement Learning","arxiv_id":"1707.08475","date":"2017-07-26","proceeding":"ICML 2017 8","authors":["Irina Higgins","Arka Pal","Andrei A. Rusu","Loic Matthey","Christopher P. Burgess","Alexander Pritzel","Matthew Botvinick","Charles Blundell","Alexander Lerchner"],"abstract":"Domain adaptation is an important open problem in deep reinforcement learning\n(RL). In many scenarios of interest data is hard to obtain, so agents may learn\na source policy in a setting where data is readily available, with the hope\nthat it generalises well to the target domain. We propose a new multi-stage RL\nagent, DARLA (DisentAngled Representation Learning Agent), which learns to see\nbefore learning to act. DARLA's vision is based on learning a disentangled\nrepresentation of the observed environment. Once DARLA can see, it is able to\nacquire source policies that are robust to many domain shifts - even with no\naccess to the target domain. DARLA significantly outperforms conventional\nbaselines in zero-shot domain adaptation scenarios, an effect that holds across\na variety of RL environments (Jaco arm, DeepMind Lab) and base RL algorithms\n(DQN, A3C and EC).","url_abs":"http://arxiv.org/abs/1707.08475v2","url_pdf":"http://arxiv.org/pdf/1707.08475v2.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":"darla-improving-zero-shot-transfer-in","repo_url":"https://github.com/BCHoagland/DARLA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[{"method_slug":"a3c","method_name":"A3C"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"entropy-regularization","method_name":"Entropy Regularization"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.08475","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}