Snake-Game-using-Deep-Reinforcement-Learning

This project aims to develop an AI model capable of learning how to play the popular game Snake from scratch using Deep Reinforcement Learning. The model starts with no prior knowledge and must learn on its own to maximize the score or reward by formulating an effective strategy. The agent receives state information about the game environment and learns to make decisions without explicit instructions. Through the Deep Q-Learning algorithm, the agent identifies optimal strategies to maximize its in-game score. The AI learns to navigate the snake to find and consume food while avoiding collisions with boundaries and its own body. As the agent successfully finds food, the snake grows. However, if the snake crashes into the boundaries or itself, the game ends. This project will teach you about the most important parts of DRL, including Q-learning, DQN architectures, experience replay, and reward engineering.