Games Research, AI, and Learning Lab
The GRAIL Lab is headed by Dr. Matthew Guzdial at the Department of Computing Science and the Alberta Machine Intelligence Institute at the University of Alberta. We are currently undertaking research in three primary research areas.
First, the application of artificial intelligence (AI) and machine learning (ML) to the generation of content, with a focus on the domain of video games, known as Procedural Content Generation.
Second, the application of AI and ML to better understanding users, players, and designers, known as User Modeling.
Third, the application of concepts from computational creativity to core AI and ML problems, particularly Transfer Learning and Explainable AI.
Note: This website is currently under construction.
Artificial Intelligence series: AI wearable technology
Artificial Intelligence series: AI wearable technology
Google AI Overview makes mistake in counting letters in word Google and name of its own company
Google AI miscounts letters, researchers explain why
Artificial Intelligence Series: how AI is impacting tech sector jobs
Artificial intelligence blame for video-game industry layoffs may be misguided
Time-based Chart Partitioning: Improving Local Coherency in Rhythm Game Chart Generation
Does anyone still clip coupons? Consumers want deals, but how they find them is shifting
Duolingo’s CEO says AI will soon replace teachers. But ... should it?
Evaluating the Effects of AI Directors for Quest Selection
Mechanic Maker: Accessible Game Development via Symbolic Learning Program Synthesis
The FarmQuest Player Telemetry Dataset: Playthrough Data of a Cozy Farming Game
Why AI can’t spell ‘strawberry’
Google DeepMind’s new generative model makes Super Mario–like games from scratch
Mechanic Maker 2.0: Reinforcement Learning for Evaluating Generated Rules
Reconstructing Existing Levels through Level Inpainting
Ask a Prof: Is AI software plagiarism?
L'Atelier
FarmQuest: A Demonstration of an AI Director Video Game Test Bed
World Models with an Entity-Based Representation
A Demonstration of KiaiTime: A Mixed-Initiative PCGML Rhythm Game Editor
The Impact of Visualizing Design Gradients for Human Designers
Image-to-Level: Generation and Repair
How Artificial Intelligence Could Help Video Gamers Create the Exact Games They Want to Play
What's in a game? Welcoming AI and creativity expert, Matthew Guzdial
An AI That Could Create Video Games by Just Watching The Gameplays
Automated Game Design via Conceptual Expansion
This Algorithm Makes Its Own Video Games
AI Learns To Program Super Mario Bros
AI learns to recreate video game after two minutes of watching someone play it
AI recreates Super Mario Bros just by watching a two-minute clip of it
New AI system can learn videogame with two minutes of footage
AI recreates Super Mario Bros. by watching gameplay videos
Artificial intelligence uses less than two minutes of videogame footage to recreate game engine
Game Level Generation from Gameplay Videos