Continuously Discovering Novel Strategies via Reward-Switching Policy Optimization
This addresses the challenge of strategy diversity in RL for applications such as game AI and robotics, though it appears incremental as it builds on existing policy optimization methods.
The paper tackles the problem of discovering diverse strategies in complex reinforcement learning environments by introducing Reward-Switching Policy Optimization (RSPO), which iteratively finds novel policies that are locally optimal and distinct from existing ones, achieving a wide spectrum of strategies across domains like MuJoCo and StarCraftII.
We present Reward-Switching Policy Optimization (RSPO), a paradigm to discover diverse strategies in complex RL environments by iteratively finding novel policies that are both locally optimal and sufficiently different from existing ones. To encourage the learning policy to consistently converge towards a previously undiscovered local optimum, RSPO switches between extrinsic and intrinsic rewards via a trajectory-based novelty measurement during the optimization process. When a sampled trajectory is sufficiently distinct, RSPO performs standard policy optimization with extrinsic rewards. For trajectories with high likelihood under existing policies, RSPO utilizes an intrinsic diversity reward to promote exploration. Experiments show that RSPO is able to discover a wide spectrum of strategies in a variety of domains, ranging from single-agent particle-world tasks and MuJoCo continuous control to multi-agent stag-hunt games and StarCraftII challenges.