Partially Observable MDPs, Monte Carlo Methods and Fishery Applications

Partially Observable Markov Decision Processes (POMDPs) provide a general mathematical framework forsequential decision making under uncertainty. However, solving POMDPs effectively under realistic assumptionsremains a challenging problem. This project aims to develop new efficient Monte Carlo algorithms to significantlyadvance the application of POMDPs to real-world decision problems involving complex action spaces and systemdynamics. Both theoretical and algorithmic approaches will be applied to sustainable fishery management --- animportant problem for Australia and an ideal context for POMDPs. The project will advance research in artificialintelligence, dynamical systems, and fishery operations, and benefit the national economy.