Tetris Q Learning Agent

this project implements a q-learning agent for playing tetris. the agent uses a neural network-based q-function to learn optimal strategies for placing tetris pieces. the implementation includes custom feature extraction, reward function design, and exploration strategies to improve learning performance