RL Final Project
Created an RL agent for 2048 game based on Temporal-Difference Learning and Multi-Stage Temporal- Difference Learning Algorithms.
Result: The agent was to able to converge to 2048 much faster than humans and showing a better performance with the Multi-Stage Temporal Learning Algorithm.