This repository contains the code for a project paper for a Master's module in the field of reinforcement learning. The aim of the project is to explore and implement Proximal Policy Optimization (PPO) agents to learn and play the 7x7 Hex game.

Automated gaming using machine learning

Java Garbage Collector Performance Tuning with Reinforcement Learning Methods

Toy model implementing various architectures to teach a generic point of mass to reach a static and/or dynamic target in a 3D space.

An RL agent that finds an optimal policy of cleaning dirt off a floor with a power washer.

Unleashing the Power of PPO: Mastering Super Mario with Reinforcement Learning. Dive into our journey of training a Proximal Policy Optimization (PPO) agent to conquer the classic NES.

Trained 2 PPO agents to dominate Pong via self-play. Achieved ~250 reward (11-0 wins, 12+ hit rallies) after 4000 episodes using Tanh activation

Flappy Bird using Reinforcement Learning

This project is based on fine-tuning LLM models (FLAN-T5) for text summarisation task using PEFT approach. All evaluation metrics being computed on ROUGE scoring and LoRA optimisation techniques being used for fine-tuning.

This project develops a reinforcement learning (RL) agent to safely navigate indoor environments, specifically designed to assist visually impaired individuals. The agent learns to avoid obstacles, utilize doorways, and efficiently reach target rooms within a simulated house layout.

An adaptive Machine Reinforcement Learning (MRL) system is being developed to gather and analyze media data using web scraping, training models to predict outcomes in areas like stock market trends, sports events, and other performance domains. It continuously refines its strategies based on real-time data and evolving patterns.

CleanRL implementation of "Mitigating Plasticity Loss in Continual Reinforcement Learning by Reducing Churn"

DQN and PPO Reinforcement Learning Agents for Snake Game (TensorFlow) 🐍

Training a Mario reinforcement learning agent using Open AI Gym and Stable Baselines 3 PPO algorithm.

Injecting object-level implicit priors in a PPO agent with contrastive learning algorithm ST-DIM.

Snake game environment integrated with OpenAI Gym. Proximal Policy Optimization (PPO) implementation for training. Visualization of training progress and agent performance. Easy to understand code.

Short own implementation of the game snake. In this project I'am using the ray library together with ray tune and a custom PPO model.

The aim of this repository is the analysis and study of computer intelligence and in-depth learning techniques in the development of intelligent gaming agents.