CTD4 -- A Deep Continuous Distributional Actor-Critic Agent with a Kalman Fusion of Multiple Critics

TL;DR

This paper introduces a novel continuous distributional reinforcement learning algorithm with an ensemble of critics fused via Kalman filtering, improving sample efficiency and reducing overestimation bias in continuous control tasks.

Contribution

It presents the first continuous distributional RL algorithm with a Kalman-fused critic ensemble, simplifying implementation and enhancing performance in continuous action spaces.

Findings

Improved sample efficiency in continuous control tasks.

Effective reduction of overestimation bias through Kalman fusion.

Competitive performance against existing methods.

Abstract

Categorical Distributional Reinforcement Learning (CDRL) has demonstrated superior sample efficiency in learning complex tasks compared to conventional Reinforcement Learning (RL) approaches. However, the practical application of CDRL is encumbered by challenging projection steps, detailed parameter tuning, and domain knowledge. This paper addresses these challenges by introducing a pioneering Continuous Distributional Model-Free RL algorithm tailored for continuous action spaces. The proposed algorithm simplifies the implementation of distributional RL, adopting an actor-critic architecture wherein the critic outputs a continuous probability distribution. Additionally, we propose an ensemble of multiple critics fused through a Kalman fusion mechanism to mitigate overestimation bias. Through a series of experiments, we validate that our proposed method provides a sample-efficient solution for executing complex continuous-control tasks.