Why Online Reinforcement Learning is Causal

arXiv:2403.04221v1 Announce Type: new
Abstract: Reinforcement learning (RL) and causal modelling naturally complement each other. The goal of causal modelling is to predict the effects of interventions in an environment, while the goal of reinforcement learning is to select interventions that maximize the rewards the agent receives from the environment. Reinforcement learning includes the two most powerful sources of information for estimating causal relationships: temporal ordering and the ability to act on an environment. This paper examines which reinforcement learning …

abstract agent arxiv cs.ai cs.lg effects environment modelling online reinforcement learning reinforcement reinforcement learning the environment type