Unified models capable of solving a wide variety of tasks have gained traction in vision and NLP due to their ability to share regularities and structures across tasks, which improves individual task performance and reduces computational footprint. However, the impact of such models remains limited in embodied learning problems, which present unique challenges due to interactivity, sample inefficiency, and sequential task presentation. In this work, we present PolyTask, a novel method for learning a single unified model that can solve various embodied tasks through a 'learn then distill' mechanism. In the 'learn' step, PolyTask leverages a few demonstrations for each task to train task-specific policies. Then, in the 'distill' step, task-specific policies are distilled into a single policy using a new distillation method called Behavior Distillation. Given a unified policy, individual task behavior can be extracted through conditioning variables. PolyTask is designed to be conceptually simple while being able to leverage well-established algorithms in RL to enable interactivity, a handful of expert demonstrations to allow for sample efficiency, and preventing interactive access to tasks during distillation to enable lifelong learning. Experiments across three simulated environment suites and a real-robot suite show that PolyTask outperforms prior state-of-the-art approaches in multi-task and lifelong learning settings by significant margins.
Score and Distribution Matching Policy: Advanced Accelerated Visuomotor Policies via Matched Distillation
Westlake University（西湖大学）；Southwest University（西南大学）；Zhejiang University（浙江大学）；Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing（北京未来区块链与隐私计算高精尖创新中心）；School of Artificial Intelligence, Beihang University（北京航空航天大学人工智能学院）
From Flow to One Step: Real-Time Multi-Modal Trajectory Policies via Implicit Maximum Likelihood Estimation-based Distribution Distillation
TAMS (Technical Aspects of Multimodal Systems), Department of Informatics, University of Hamburg（汉堡大学信息学院TAMS（多模态系统技术方面））；Technical University of Munich（慕尼黑技术大学）；Agile Robots SE（敏捷机器人公司）
StyleLoco: Generative Adversarial Distillation for Natural Humanoid Robot Locomotion
National Key Laboratory of General Artificial Intelligence, BIGAI（通用人工智能国家重点实验室，BIGAI）；School of Control Science and Engineering, Shandong University（山东大学控制科学与工程学院）；Huazhong University of Science and Technology（华中科技大学）
UniLegs: Universal Multi-Legged Robot Control through Morphology-Agnostic Policy Distillation
Wuxi Research Institute of Applied Technologies, Tsinghua University（清华大学应用技术研究院（无锡））；School of Internet of Things Engineering, Department of Automation, Jiangnan University（江南大学物联网工程学院）；Shanghai Jiao Tong University（上海交通大学）；School of Automation Science and Electrical Engineering, Beihang University（北航自动化科学与电气工程学院）
