Apply AI techniques in games and 3D applications
This course shows you how modern AI techniques can be used to build smarter, more dynamic games and interactive 3D applications. You will explore how AI can “play games” or “control characters”, generate content (like levels, environments, or assets), and influence player experience, using tools like Python-based AI frameworks, Unity (C#) or the Unreal Game Engine (C++), though you are free to choose other frameworks if you prefer.
You will learn about classic machine learning methods like genetic algorithms, reinforcement learning, and tree search, and gain the technical skills to design and implement intermediate- to advanced-level AI/procedural content generation (PCG) systems. Further, you will learn how advantages and disadvantages of current state-of-the-art AI models and how they can be embedded in your own applications.
The course combines theory and practice roughly equally: some classes introduce AI concepts, while the practical work happens across tutorials and a research group project. There, you define your own research question for applying AI to a game context and build it, from planning to coding.
By the end of the course, you will be able to choose suitable AI methods for a given task, program them, integrate them into a game or simulation, and assess whether your system works as intended.
What You Will Learn:
- AI Methods, i.e., (Deep) Reinforcement Learning, Neural Networks,
- Evolutionary Algorithms
- AI Models, i.e., Large Language Models, Image/Video/Sound Generation Models
- Application Areas, e.g., Procedural Content Generation, AI Decision-Making
Course Overview:
~5 ECTS theory (AI methods + game applications)
~5 ECTS group project on a self-defined game research AI/PCG project
Oral group examination based on project report
