Paper ID: 2409.04374

Gaussian-Mixture-Model Q-Functions for Reinforcement Learning by Riemannian Optimization

Minh Vu, Konstantinos Slavakis

This paper establishes a novel role for Gaussian-mixture models (GMMs) as functional approximators of Q-function losses in reinforcement learning (RL). Unlike the existing RL literature, where GMMs play their typical role as estimates of probability density functions, GMMs approximate here Q-function losses. The new Q-function approximators, coined GMM-QFs, are incorporated in Bellman residuals to promote a Riemannian-optimization task as a novel policy-evaluation step in standard policy-iteration schemes. The paper demonstrates how the hyperparameters (means and covariance matrices) of the Gaussian kernels are learned from the data, opening thus the door of RL to the powerful toolbox of Riemannian optimization. Numerical tests show that with no use of training data, the proposed design outperforms state-of-the-art methods, even deep Q-networks which use training data, on benchmark RL tasks.

Submitted: Sep 6, 2024