This project implements a Deep Reinforcement Learning agent that learns to play Flappy Bird using the Deep Q-Network (DQN) algorithm built with PyTorch and Gymnasium. The agent learns through trial and error by interacting with the environment, storing experiences in a replay buffer, and updating its policy using the Bellman equation.