Uncertainty Quantification and Exploration for Reinforcement Learning
About
We investigate statistical uncertainty quantification for reinforcement learning (RL) and its implications in exploration policy. Despite ever-growing literature on RL applications, fundamental questions about inference and error quantification, such as large-sample behaviors, appear to remain quite open. In this paper, we fill in the literature gap by studying the central limit theorem behaviors of estimated Q-values and value functions under various RL settings. In particular, we explicitly identify closed-form expressions of the asymptotic variances, which allow us to efficiently construct asymptotically valid confidence regions for key RL quantities. Furthermore, we utilize these asymptotic expressions to design an effective exploration strategy, which we call Q-value-based Optimal Computing Budget Allocation (Q-OCBA). The policy relies on maximizing the relative discrepancies among the Q-value estimates. Numerical experiments show superior performances of our exploration strategy than other benchmark policies.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Empirical Coverage Estimation | RiverSwim | Q^π(1, 0)0.948 | 120 | |
| Optimal Policy Selection | Stochastic Gridworld (S=16, A=4) | PCS92 | 28 | |
| Action-Value coverage estimation | RiverSwim mostly-right target policy T=50 | Q-Value Estimate (s=1, a=0)0.503 | 20 | |
| State Value Estimation Coverage | RiverSwim | Value Estimate State 10.949 | 20 | |
| State-Action Value Estimation Coverage | RiverSwim | Q-Value Estimate (s=1, a=0)0.949 | 20 | |
| State-Value coverage estimation | RiverSwim mostly-right target policy T=50 | V(s=1)0.503 | 20 | |
| Empirical Coverage Estimation | RiverSwim episode length T = 10 (nominal 95% coverage) | Q* (1, 0)38.8 | 20 | |
| Off-policy Evaluation | RiverSwim mostly-left policy, T=50 | Qπ(1, 0) Coverage52.3 | 20 | |
| Empirical Coverage Estimation | RiverSwim T=50 90% nominal coverage | Q* (1, 0)47.8 | 20 | |
| Optimal Policy Recovery (Empirical Coverage) | RiverSwim T=50 nominal 95% coverage | Q* Recovery (s=1, a=0)49.7 | 20 |