Terence Broad

Terence Broad

Machine Learning for Game World Generation using Intrinsic Motivation as an Objective Function

New forms of machine learning are increasingly important in games research, and generative models demonstrate great potential for content creation. Convolutional Neural Networks can be used for texture synthesis, new forms of rendering, model generation and animation. When such systems are computationally efficient enough to render in real time, they will transform the way video games are both produced and played.

Current deep learning systems can be very successful at performing inference on the statistical distribution of high dimensional data. But they are not very good at novel content generation. Jurgen Schmidhuber has proposed (in his 2008 paper “Driven by Compression Progress”) a Reinforcement Learning agent, motivated not by external reward, but by intrinsic motivation. The agent constantly samples the world, trying to compress it, inferring regularities in the data, but motivated to seek out data that is novel. Developing over time more sophisticated understanding of the world. In this PhD I will develop new kinds of machine learning agents driven by intrinsic motivation, for the purposes of content generation in games, in particular towards procedural content generation systems that display complex emergent behaviour.

Terence Broad is a researcher and artist based in London. He completed his Msci in Creative Computing at Goldsmiths in 2016 where his research interests were in generative machine learning systems and computation photography. In the following two years he worked at the tech startup Vivacity Labs where he was responsible for researching and deploying deep learning based computer vision models on embedded IoT sensors at city-wide scales. His artwork has been shown internationally, at venues including The Barbican, Ars Electronica and The Whitney Museum of American Art.

Home institution: Goldsmiths

Supervisor: Dr Mick Grierson

Ready to apply?

Once you have identified your potential supervisor, we would encourage you to contact them to discuss your research proposal.

Learn More