Recreating Bill Seiler's 1985 version of Space War and training RL agents with Self-Play

The game utilizes the Bellman equation as a fundamental component in its algorithm, incorporating libraries such as NumPy and PyTorch to efficiently perform mathematical computations and enable it not only to make decisions during gameplay but also to continuously train and improve itself through iterative learning processes.