Hierarchically Decoupled Imitation for Morphological Transfer
arXiv:2003.01709
Abstract
Learning long-range behaviors on complex high-dimensional agents is a fundamental problem in robot learning. For such tasks, we argue that transferring learned information from a morphologically simpler agent can massively improve the sample efficiency of a more complex one. To this end, we propose a hierarchical decoupling of policies into two parts: an independently learned low-level policy and a transferable high-level policy. To remedy poor transfer performance due to mismatch in morphologies, we contribute two key ideas. First, we show that incentivizing a complex agent's low-level to imitate a simpler agent's low-level significantly improves zero-shot high-level transfer. Second, we show that KL-regularized training of the high level stabilizes learning and prevents mode-collapse. Finally, on a suite of publicly released navigation and manipulation environments, we demonstrate the applicability of hierarchical transfer on long-range tasks across morphologies. Our code and videos can be found at https://sites.google.com/berkeley.edu/morphology-transfer.
arXiv commentInternational Conference on Machine Learning (ICML) 2020 camera ready submission
References in corpus (15)
- Continuous control with deep reinforcement learning
- Trust Region Policy Optimization
- Soft Actor-Critic Algorithms and Applications
- Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
- Emergent Tool Use From Multi-Agent Autocurricula
- Modular Multitask Reinforcement Learning with Policy Sketches
- Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning
- Deep Decentralized Multi-task Multi-Agent Reinforcement Learning under Partial Observability
- Third-Person Imitation Learning
- Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity
- Near-Optimal Representation Learning for Hierarchical Reinforcement Learning
- Hierarchical Reinforcement Learning with Advantage-Based Auxiliary Rewards
- Hardware Conditioned Policies for Multi-Robot Transfer Learning
- Third-Person Visual Imitation Learning via Decoupled Hierarchical Controller
- Skill Transfer in Deep Reinforcement Learning under Morphological Heterogeneity