Machine Learning Transfer Research: Arcade Prototype

Arcade prototype for the research project: Evaluating transfer learning for reinforcement learning agents across same-genre video games
Research project carried out in partnership between PulluP Entertainment and ISART Digital Paris

Table of Contents

• Presentation
• Setup
• Technology
• Credits

Presentation

This project was developed in Unity by a team of four to prototype an arcade racing game in order to train on it, an agent with machine learning and transfer it to the other prototype we made, a Realistic prototype.

Setup

Warning: Make sure to have installed the python environment

Start a training session

In Anaconda Prompt or MiniForge Prompt

1. Go to the project folder, then to the Assets/ML-Agents folder
2. Activate the python environment

conda activate mlagents

3. Start training (replace XXX by the training number, e.g., 005)

mlagents-learn Config/trainer_config.yaml --run-id=trainingXXX

In Unity

1. Open the scenes Assets/scenes/Training
2. Click on the play button

Test trained agents

In Unity

1. Open the scene Assets/scenes/Testing
2. In the Hierarchy Car -> MovementRigibody GameObject
3. In the Inspector Behavior Parameters Script
4. Change the parameters Model with the neural network you want to test. Make sure that Behavior Type is set to Inference Only
5. Launch the scene

Evaluate trained agent

In Unity

1. Open the scenes Assets/scenes/Testing
2. In the Hierarchy EnvTraining GameObject
3. In the Inspector Evaluation Tests Script
4. Set the parameters Model Tested with the neural network you want to evaluate
5. Launch the scene

Technology

• Unity 6 v6000.3.9f1
• ML Agents Plugin (version 4.0.2)

Credits

• Bryan BACHELET
• Vincent DEVINE
• Matéo ERBISTI
• Omaya LISE
• Aurelien CHAMBON
• Aurélien LHERBIER

Assets

• Dreamteck Spline by Dreamteck
• Rally cars by Ash Dev
• Racing Kit by Kenny