This is my master thesis project, with the goal of exploring how VR can be utilized as a tool for learning in AI education. I developed a prototype application for the Oculus Quest, where students are given an introduction to deep learning and neural networks. Users find themselves in an escape room environment, where topics have been split into separate rooms. The user progress through the application by doing 3D-puzzles, calculations, and quizzes, based on the course-material.
There is a large need for AI competence worldwide. Having good technological tools for learning has shown to be more important than ever, due to the global pandemic of 2020. We wanted to see if VR can be a valuable tool for fulfulling the need of AI competence.
A link to the master thesis report will be added here when it is published. We are also planning to write a research paper about the project.
This 5-minute YouTube video explains the core-concepts and shows what was developed for the application.
The image to the left shows one of the tasks where the user learns neural network notation by placing neurons in a neural network.
The image to the right shows the visualization of gradient descent.
Since many users in the target audience are assumed to be new to VR, they have to play through a 5-10 minute tutorial, where every interaction needed is taught. Tutorial play-through video.
The curriculum was split into separate rooms. The user needs to collect and win cartridges loaded with quizzes for each topic in order to complete the application. Playing through the application takes approximately 30-90 minutes. Deep learning introduction play-through video. The topics covered in the application are:
- Neurons
- Cost functions
- Gradient Descent
- Backpropagation
The technologies listed below were used for developing the application for the Oculus Quest. I used inspiration from the Design, Develop, and Deploy for VR Unity Learn course by Oculus for setting up the fundamentals. This project supports the Oculus Quest, Rift, and Rift S, but it should be possible to port it to OpenVR devices without too much effort.
Unity 2018 LTS version.
- Oculus Integration for hardware integration.
- Virtual Reality Toolkit v4 (VRTK) for interactions.
- JSON .NET For Unity for handling JSON files in Unity.
- Unity Snaps for efficiently building room layouts.
- Download or clone the project.
- Open Unity Hub and add the project.
- Make sure you have the right version of Unity installed.
- Open the project.
If you want to try the application, you can find it in this Google Drive for Oculus Quest, Rift, and Rift S. Installation guides are provided for all devices.
The application was developed by Sølve Robert Bø Hunvik during his master thesis project at the Norwegian University of Science and Technology in the Department of Computer Science. Frank Lindseth was the project's supervisor. Every free resource used for the project is credited within the game menu.
The project is licensed under the MIT license. More details in LICENSE.md.
Copyright © 2020 Sølve Robert Bø Hunvik