Observation Space Matters: Benchmark and Optimization Algorithm
This addresses the problem of inefficient observation space design in deep RL for researchers and practitioners, offering an incremental optimization method.
The paper tackled the sensitivity of deep reinforcement learning to observation space design by benchmarking common choices and proposing a search algorithm to optimize them, resulting in significantly improved learning speed compared to manual designs.
Recent advances in deep reinforcement learning (deep RL) enable researchers to solve challenging control problems, from simulated environments to real-world robotic tasks. However, deep RL algorithms are known to be sensitive to the problem formulation, including observation spaces, action spaces, and reward functions. There exist numerous choices for observation spaces but they are often designed solely based on prior knowledge due to the lack of established principles. In this work, we conduct benchmark experiments to verify common design choices for observation spaces, such as Cartesian transformation, binary contact flags, a short history, or global positions. Then we propose a search algorithm to find the optimal observation spaces, which examines various candidate observation spaces and removes unnecessary observation channels with a Dropout-Permutation test. We demonstrate that our algorithm significantly improves learning speed compared to manually designed observation spaces. We also analyze the proposed algorithm by evaluating different hyperparameters.