Neural Fitted Q Iteration - First Experiences with a Data Efficient Neural Reinforcement Learning Method

This project aims to develop algorithms using linear programming to optimise the dispatch behaviour of a battery based in Victoria.