Observation Space Matters: Benchmark and Optimization Algorithm
Joanne Taery Kim,Sehoon Ha,Joanne Taery Kim,Sehoon Ha
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...
