Welcome to the Snake AI project! This project focuses on implementing an artificial intelligence system that gradually learns to play the classic Snake Game using a Linear QNet.

The Snake AI utilizes Reinforcement Learning techniques, specifically the Q-learning algorithm, to train an artificial agent to play the Snake Game. The QNet is implemented in a linear fashion to simplify the learning process.

To run the Snake AI project, follow these steps:

1. Clone the repository:

   git clone https://github.com/anshumalivfx/AI_Learning_Snake_Game_LinearQNet.git

2. Install the required dependencies:

   pip install -r requirements.txt

3. Run the Snake AI program:

   python agent.py

The core of the learning mechanism lies in the Linear QNet. It is a simplified neural network that represents the Q-function, which maps state-action pairs to their respective Q-values. The mathematical representation is as follows:

$$ Q(s, a) = W \cdot X $$

where:

• ( Q(s, a) ) is the Q-value for a given state-action pair.
• ( W ) is the weight vector.
• ( X ) is the feature vector representing the state-action pair.

The Q-learning algorithm is used to update the Q-values based on the rewards received by the agent. The update equation is given by:

$$ Q(s, a) = Q(s, a) + \alpha \cdot [R + \gamma \cdot \max(Q(s', a')) - Q(s, a)] $$

where:

• α is the learning rate.
• γ is the discount factor.
• ( R ) is the immediate reward.
• ( s ) and ( s' ) are the current and next states, respectively.
• ( a ) and ( a' ) are the current and next actions, respectively.

The Snake AI is trained by allowing the agent to play the Snake Game multiple times, and the Q-values are updated using the Q-learning algorithm. Over time, the agent learns a policy that maximizes its cumulative rewards.

Feel free to tweak the hyperparameters in the config.py file to experiment with different settings and improve the learning performance.

# Example hyperparameters in config.py
learning_rate = 0.001
gamma = 0
epsilon = 0

Contributions are welcome! If you have any ideas for improvements or bug fixes, please open an issue or submit a pull request.

Happy coding! 🐍🤖