Introductory description

This module introduces the principles and practical applications of artificial intelligence within games. Game AI has historically been a major driver of innovation in interactive systems, combining algorithmic techniques with data-driven approaches to create believable agents, intelligent behaviours, and adaptive game systems. As modern games increasingly integrate machine learning techniques alongside traditional AI approaches, it is important that graduates understand both the classical foundations and contemporary developments in this field.

The module is structured around two complementary themes: traditional game AI techniques and modern machine learning methods used in games. Traditional techniques include pathfinding algorithms, behaviour trees, and decision systems, which form the backbone of many interactive and real-time gameplay systems. These approaches emphasise deterministic reasoning, efficient algorithms, and the design of robust decision-making systems suitable for real-time environments.

Alongside these foundations, the module introduces key concepts from modern artificial intelligence and machine learning. Students will explore neural networks and their evolution from simple architectures to more advanced models such as convolutional neural networks, recurrent networks, and transformers. The module also examines how these techniques are applied within the games industry, including AI agents, procedural content generation, rendering enhancement and denoising, physics approximation, and generative systems for textures, models, and dialogue.

The emphasis throughout the module is on understanding core principles and applying them in practice. Through this combination of theory and implementation, the module aims to provide students with both conceptual understanding and practical skills relevant to modern game development and interactive AI systems.

Module aims

Outline syllabus

This is an indicative module outline only to give an indication of the sort of topics that may be covered. Actual sessions held may differ.

Introduction to Artificial Intelligence in Games
Search and Pathfinding
Decision Making Systems for Game Agents
Foundations of Machine Learning
Neural Network Architectures
AI Agents and Learning in Games
AI for Game Content and Systems
Modern AI Applications in Games

Learning outcomes

By the end of the module, students should be able to:

• Critically analyse the principles and techniques used in traditional game artificial intelligence, including pathfinding, decision systems, and behaviour modelling.
• Evaluate and compare different AI approaches used in games, including rule-based systems and machine learning methods, and justify their suitability for specific design or technical problems.
• Design and implement classical game AI algorithms within an interactive environment, such as pathfinding or behaviour systems for game agents.
• Demonstrate a conceptual understanding of modern machine learning techniques, including neural networks, convolutional networks, recurrent networks, and transformer architectures.
• Develop and apply neural network–based solutions to solve practical problems relevant to games or interactive systems.
• Critically assess the role and impact of modern AI technologies in game development, including applications such as agent behaviour, content generation, rendering enhancement, and large language model integration.

Subject specific skills

Mathematical skills, programming skills, AI skills

Transferable skills

Technology literacy, adaptability