Documentation

This repo is mostly designed based on Unity ML-Agents. Please first make sure you at least already know how to make a learning environment in ML-Agens.

For ML-Agents or related machine learning knowledge, see ML-Agents documentation.

Installation & Set-up

• Installation

Getting Started

• Getting Started with the 3D Balance Ball Environment
• Features and Basic Concepts
• Example Environments

Learning

• Training with Proximal Policy Optimization(PPO)
• Training with Imitation(Supervised Learning)
• Use Neural Evolution to optimize Neural Network
• Training a Generative Adversarial Network(GAN)

MAES Optimization

• Use MAES Optimization to Find the Best Solution

Others

• Define Your Own Training Process for ML-Agent - See the source codes of Trainer.cs for details.