Nonstationary Reinforcement Learning with Linear Function Approximation

Abstract

We consider reinforcement learning (RL) in episodic Markov decision processes (MDPs) with linear function approximation under drifting environment. Specifically, both the reward and state transition functions can evolve over time but their total variations do not exceed a $\text{variation budget}$. We first develop $\text{LSVI-UCB-Restart}$ algorithm, an optimistic modification of least-squares value iteration with periodic restart, and bound its dynamic regret when variation budgets are known. Then we propose a parameter-free algorithm $\text{Ada-LSVI-UCB-Restart}$ that extends to unknown variation budgets. We also derive the first minimax dynamic regret lower bound for nonstationary linear MDPs and as a byproduct establish a minimax regret lower bound for linear MDPs unsolved by Jin et al. (2020). Finally, we provide numerical experiments to demonstrate the effectiveness of our proposed algorithms.

References in corpus (18)

• Solving Rubik's Cube with a Robot Hand
• Safe, Multi-Agent, Reinforcement Learning for Autonomous Driving
• Provably Efficient Reinforcement Learning with Linear Function Approximation
• Deep Exploration via Randomized Value Functions
• Contextual Decision Processes with Low Bellman Rank are PAC-Learnable
• Learning Near Optimal Policies with Low Inherent Bellman Error
• Model-based RL in Contextual Decision Processes: PAC bounds and Exponential Improvements over Model-free Approaches
• Provably efficient RL with Rich Observations via Latent State Decoding
• On Lower Bounds for Regret in Reinforcement Learning
• FLAMBE: Structural Complexity and Representation Learning of Low Rank MDPs
• Hedging the Drift: Learning to Optimize under Non-Stationarity
• A Simple Approach for Non-stationary Linear Bandits
• Reinforcement Learning with General Value Function Approximation: Provably Efficient Approach via Bounded Eluder Dimension
• Variational Regret Bounds for Reinforcement Learning
• Dynamic Regret of Policy Optimization in Non-stationary Environments
• Online learning in MDPs with linear function approximation and bandit feedback
• Nearly Optimal Algorithms for Piecewise-Stationary Cascading Bandits
• Learning Infinite-horizon Average-reward MDPs with Linear Function Approximation

Cited by in corpus (3)

• Non-stationary Reinforcement Learning without Prior Knowledge: An Optimal Black-box Approach
• Efficient Learning in Non-Stationary Linear Markov Decision Processes
• Optimistic Policy Optimization is Provably Efficient in Non-stationary MDPs