Reinforcement learning: Implementing Ada-FP Algorithm

Project for the M2 Math and AI reinforcement learning course.

The aim of the project is to implement and test the Ada FP algorithm. The theoretical explanation is available in the report.pdf, while the notebook contains the code used for the algorithm.

The notebook.ipynb contains the calculation for the value function with classical iteration with $B_*$ and $B_\pi$. It was also tried with the Ada-FP algorithm, Ada-FP component per component and for iteration over the action-value function.