Bayesian policy selection using active inference
This work addresses robustness and efficiency issues in AI agents, but it is incremental as it adapts an existing theory to a known benchmark.
The paper tackles the problem of sample inefficiency and generalization in reinforcement learning by applying active inference, a neuroscience theory, to the mountain car problem, showing it can unify RL and learning from demonstrations.
Learning to take actions based on observations is a core requirement for artificial agents to be able to be successful and robust at their task. Reinforcement Learning (RL) is a well-known technique for learning such policies. However, current RL algorithms often have to deal with reward shaping, have difficulties generalizing to other environments and are most often sample inefficient. In this paper, we explore active inference and the free energy principle, a normative theory from neuroscience that explains how self-organizing biological systems operate by maintaining a model of the world and casting action selection as an inference problem. We apply this concept to a typical problem known to the RL community, the mountain car problem, and show how active inference encompasses both RL and learning from demonstrations.