Online Learning and Planning in Partially Observable Domains without Prior Knowledge
This work addresses the problem of planning in partially observable environments for AI agents, offering an incremental improvement by enabling online learning without prior knowledge.
The paper tackles the challenge of optimal action in stochastic, partially observable domains by proposing an online model-based planning approach using Predictive State Representations, which eliminates the need for prior knowledge and achieves high performance compared to state-of-the-art methods.
How an agent can act optimally in stochastic, partially observable domains is a challenge problem, the standard approach to address this issue is to learn the domain model firstly and then based on the learned model to find the (near) optimal policy. However, offline learning the model often needs to store the entire training data and cannot utilize the data generated in the planning phase. Furthermore, current research usually assumes the learned model is accurate or presupposes knowledge of the nature of the unobservable part of the world. In this paper, for systems with discrete settings, with the benefits of Predictive State Representations~(PSRs), a model-based planning approach is proposed where the learning and planning phases can both be executed online and no prior knowledge of the underlying system is required. Experimental results show compared to the state-of-the-art approaches, our algorithm achieved a high level of performance with no prior knowledge provided, along with theoretical advantages of PSRs. Source code is available at https://github.com/DMU-XMU/PSR-MCTS-Online.