Nimble Quest AI Bot (PPO)

This repository contains a Reinforcement Learning agent trained using Proximal Policy Optimization (PPO) to play the game Nimble Quest. The agent learns to navigate and overcome challenges in the game environment.

Overview

The Nimble Quest AI Bot is built using the Stable Baselines library and trained using the PPO algorithm. It utilizes a Convolutional Neural Network (CNN) policy to process game observations and make decisions on actions to take.

Agent Gameplay

Check out the video below to see the trained agent in action:

Nimble Quest AI Bot Youtube Gameplay (clicking the image below leads to a Youtube page)

Future Improvements

• Fine-tune the hyperparameters for better performance
• Experiment with different policy networks
• Implement additional reward shaping techniques