Recent works in Reinforcement Learning (RL) combine model-free (Mf)-RL algorithms with model-based (Mb)-RL approaches to get the best from both: asymptotic performance of Mf-RL and high sample-efficiency of Mb-RL. Inspired by these works, we propose a hierarchical framework that integrates online learning for the Mb-trajectory optimization with off-policy methods for the Mf-RL. In particular, two loops are proposed, where the Dynamic Mirror Descent based Model Predictive Control (DMD-MPC) is used as the inner loop Mb-RL to obtain an optimal sequence of actions. These actions are in turn used to significantly accelerate the outer loop Mf-RL. We show that our formulation is generic for a broad class of MPC based policies and objectives, and includes some of the well-known Mb-Mf approaches. We finally introduce a new algorithm: Mirror-Descent Model Predictive RL (M-DeMoRL), which uses Cross-Entropy Method (CEM) with elite fractions for the inner loop. Our experiments show faster convergence of the proposed hierarchical approach on benchmark MuJoCo tasks. We also demonstrate hardware training for trajectory tracking in a 2R leg, and hardware transfer for robust walking in a quadruped. We show that the inner-loop Mb-RL significantly decreases the number of training iterations required in the real system, thereby validating the proposed approach.
Utkarsh A Mishra (Indian Institute of Technology, Roorkee)
Soumya Samineni (Indian Institute of Science, Bangalore)
Shalabh Bhatnagar (Indian Institute of Science)
Shalabh Bhatnagar is a Professor in the Department of Computer Science and Automation at the Indian Institute of Science, Bangalore. His research interests are in stochastic approximation, stochastic optimisation, reinforcement learning as well as applications in various engineering domains.
Shishir N Y (Indian Institute of Science)
More from the Same Authors
2021 : Dynamic Mirror Descent based Model Predictive Control for Accelerating Robot Learning »
Utkarsh A Mishra · Soumya Samineni · Aditya Varma Sagi · Shalabh Bhatnagar · Shishir N Y
2022 Poster: Model-based Safe Deep Reinforcement Learning via a Constrained Proximal Policy Optimization Algorithm »
Ashish K Jayant · Shalabh Bhatnagar
2021 : Learning Representations for Pixel-based Control: What Matters and Why? »
Manan Tomar · Utkarsh A Mishra · Amy Zhang · Matthew Taylor