Prediction-Guided Multi-Objective Reinforcement Learning for Continuous Robot Control


Many real-world control problems involve conflicting objectives where we desire a dense and high-quality set of control policies that are optimal for different objective preferences (called Pareto-optimal). While extensive research in multi-objective reinforcement learning (MORL) has been conducted to tackle such problems, multi-objective optimization for complex continuous robot control is still under-explored. In this work, we propose an efficient evolutionary learning algorithm to find the Pareto set approximation for continuous robot control problems, by extending a state-of-the-art RL algorithm and presenting a novel prediction model to guide the learning process. In addition to efficiently discovering the individual policies on the Pareto front, we construct a continuous set of Pareto-optimal solutions by Pareto analysis and interpolation. Furthermore, we design six multi-objective RL environments with continuous action space, which is the first benchmark platform to evaluate MORL algorithms on various robot control problems. We test the previous methods on the proposed benchmark problems, and the experiments show that our approach is able to find a much denser and higher-quality set of Pareto policies than the existing algorithms.

ICML 2020


A SOTA algorithm for multi-objective RL which achieves the best convergence and density of the Pareto front. Our algorithm relies on a prediction model that predicts the future improvements of the objectives and is trained using history data.