Simple Policy Optimization
This work addresses a key bottleneck in reinforcement learning for practitioners by offering a more efficient and theoretically sound method, though it is incremental as it builds on existing approaches.
The paper tackles the trade-off between theoretical robustness and practical efficiency in policy optimization by introducing Simple Policy Optimization (SPO), which modifies PPO's loss to combine strengths from TRPO and PPO, resulting in improved performance for training large networks.
Model-free reinforcement learning algorithms have seen remarkable progress, but key challenges remain. Trust Region Policy Optimization (TRPO) is known for ensuring monotonic policy improvement through conservative updates within a trust region, backed by strong theoretical guarantees. However, its reliance on complex second-order optimization limits its practical efficiency. Proximal Policy Optimization (PPO) addresses this by simplifying TRPO's approach using ratio clipping, improving efficiency but sacrificing some theoretical robustness. This raises a natural question: Can we combine the strengths of both methods? In this paper, we introduce Simple Policy Optimization (SPO), a novel unconstrained first-order algorithm. By slightly modifying the policy loss used in PPO, SPO can achieve the best of both worlds. Our new objective improves upon ratio clipping, offering stronger theoretical properties and better constraining the probability ratio within the trust region. Empirical results demonstrate that SPO outperforms PPO with a simple implementation, particularly for training large, complex network architectures end-to-end.