AIJun 4, 2017

Actor-Critic for Linearly-Solvable Continuous MDP with Partially Known Dynamics

arXiv:1706.01077v15 citations
Originality Incremental advance
AI Analysis

This addresses robotic applications where system dynamics are partially known, offering an incremental improvement over existing methods for L-MDPs.

The paper tackles reinforcement learning for continuous MDPs with partially known dynamics by proposing an actor-critic method that learns without a model of uncontrolled dynamics or transition noise, requiring only control dynamics knowledge. It shows improved learning and policy performance on synthetic and real-world traffic simulation tests.

In many robotic applications, some aspects of the system dynamics can be modeled accurately while others are difficult to obtain or model. We present a novel reinforcement learning (RL) method for continuous state and action spaces that learns with partial knowledge of the system and without active exploration. It solves linearly-solvable Markov decision processes (L-MDPs), which are well suited for continuous state and action spaces, based on an actor-critic architecture. Compared to previous RL methods for L-MDPs and path integral methods which are model based, the actor-critic learning does not need a model of the uncontrolled dynamics and, importantly, transition noise levels; however, it requires knowing the control dynamics for the problem. We evaluate our method on two synthetic test problems, and one real-world problem in simulation and using real traffic data. Our experiments demonstrate improved learning and policy performance.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes