Mean-Semivariance Policy Optimization via Risk-Averse Reinforcement Learning

Abstract

Keeping risk under control is often more crucial than maximizing expected rewards in real-world decision-making situations, such as finance, robotics, autonomous driving, etc. The most natural choice of risk measures is variance, which penalizes the upside volatility as much as the downside part. Instead, the (downside) semivariance, which captures the negative deviation of a random variable under its mean, is more suitable for risk-averse proposes. This paper aims at optimizing the mean-semivariance (MSV) criterion in reinforcement learning w.r.t. steady reward distribution. Since semivariance is time-inconsistent and does not satisfy the standard Bellman equation, the traditional dynamic programming methods are inapplicable to MSV problems directly. To tackle this challenge, we resort to Perturbation Analysis (PA) theory and establish the performance difference formula for MSV. We reveal that the MSV problem can be solved by iteratively solving a sequence of RL problems with a policy-dependent reward function. Further, we propose two on-policy algorithms based on the policy gradient theory and the trust region method. Finally, we conduct diverse experiments from simple bandit problems to continuous control tasks in MuJoCo, which demonstrate the effectiveness of our proposed methods.

Accecpted by Journal of Artificial Intelligence Research

References in corpus (9)

• Challenges of Real-World Reinforcement Learning
• Policy Gradients with Variance Related Risk Criteria
• Risk-Sensitive and Robust Decision-Making: a CVaR Optimization Approach
• Deep Dynamics Models for Learning Dexterous Manipulation
• Distributional Reinforcement Learning for Efficient Exploration
• On-Policy Deep Reinforcement Learning for the Average-Reward Criterion
• Risk-Sensitive Reinforcement Learning: Near-Optimal Risk-Sample Tradeoff in Regret
• Average-Reward Reinforcement Learning with Trust Region Methods
• A unified algorithm framework for mean-variance optimization in discounted Markov decision processes