Latent Plans for Task-Agnostic Offline Reinforcement Learning
This addresses the problem of enabling robots to perform complex, multi-step tasks from high-dimensional observations without task-specific training, representing a significant advance rather than an incremental improvement.
The paper tackles the challenge of learning long-horizon, task-agnostic policies for robot control from offline data by combining imitation learning and offline reinforcement learning in a hierarchical approach, achieving an order-of-magnitude performance improvement over state-of-the-art baselines and outperforming existing methods on 25 real-world manipulation tasks.
Everyday tasks of long-horizon and comprising a sequence of multiple implicit subtasks still impose a major challenge in offline robot control. While a number of prior methods aimed to address this setting with variants of imitation and offline reinforcement learning, the learned behavior is typically narrow and often struggles to reach configurable long-horizon goals. As both paradigms have complementary strengths and weaknesses, we propose a novel hierarchical approach that combines the strengths of both methods to learn task-agnostic long-horizon policies from high-dimensional camera observations. Concretely, we combine a low-level policy that learns latent skills via imitation learning and a high-level policy learned from offline reinforcement learning for skill-chaining the latent behavior priors. Experiments in various simulated and real robot control tasks show that our formulation enables producing previously unseen combinations of skills to reach temporally extended goals by "stitching" together latent skills through goal chaining with an order-of-magnitude improvement in performance upon state-of-the-art baselines. We even learn one multi-task visuomotor policy for 25 distinct manipulation tasks in the real world which outperforms both imitation learning and offline reinforcement learning techniques.