REValueD: Regularised Ensemble Value-Decomposition for Factorisable Markov Decision Processes

Discrete-action reinforcement learning algorithms often falter in tasks with highdimensional discrete action spaces due to the vast number of possible actions. A recent advancement leverages value-decomposition, a concept from multi-agent reinforcement learning, to tackle this challenge. This study delves deep into the effects of this value-decomposition, revealing that whilst it curtails the overestimation bias inherent to Q-learning algorithms, it amplifies target variance. To counteract this, we present an ensemble of critics to mitigate target variance. Moreover, we introduce a regularisation loss that helps to mitigate the effects that exploratory actions in one dimension can have on the value of optimal actions in other dimensions. Our novel algorithm, REValueD, tested on discretised versions of the DeepMind Control Suite tasks, showcases superior performance, especially in the challenging humanoid and dog tasks. We further dissect the factors influencing REValueD's performance, evaluating the significance of the regularisation loss and the scalability of REValueD with increasing sub-actions per dimension. N i=1 n i atomic actions. Due to the combinatorial explosion of atomic actions that must be accounted for, standard algorithms such as Q-learning (Watkins and Dayan, 1992; Mnih et al., 2013) fail to learn in these settings as a result of computational impracticalities.