The primary objective is to explore how reinforcement learning agents can learn complex spatial reasoning and planning tasks in a constrained environment. The 2x2 Rubik's Cube provides an ideal ...