Observation Space Matters: Benchmark and Optimization Algorithm

TL;DR

This paper investigates how the choice of observation spaces affects deep reinforcement learning performance, proposing an automated search algorithm to optimize observation design and improve learning efficiency.

Contribution

The paper introduces a novel search algorithm that automatically identifies optimal observation spaces for deep RL, outperforming manual design choices.

Findings

Automated search significantly improves learning speed.

Common observation design choices are verified through benchmark experiments.

The algorithm effectively removes unnecessary observation channels.

Abstract

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.