Improbable sponsors Compass PhD student in new partnership

The University of Bristol is announcing a new industrial sponsor of Compass – the EPSRC Centre for Doctoral Training in Computational Statistics and Data Science. Improbable, a global technology company which provides innovative products and services to makers of virtual worlds and simulations, is sponsoring a PhD research project entitled Agent-based model calibration using likelihood-free inference. The project’s aim is to devise a general framework for calibrating agent-based models from training data by inferring the model parameters in a statistical framework.

Founded in 2012 and headquartered in London, Improbable employs more than 700 staff across the world.

Improbable’s games business provides better ways to make multiplayer games and helps multiplayer developers to meet any challenge. Its services include managed hosting & orchestration, networking, online services and development tools, as well as advice, support and full co-development. Improbable also makes innovative multiplayer titles using its own technology.

Improbable’s Defence business operates globally and combines its parent company’s software engineering experience with expertise in computational modelling, AI and data analytics. Its work focuses on adapting and extending SpatialOS, Improbable’s multiplayer game networking solution, to enable the most sophisticated military simulations and synthetic environments ever experienced.

About working with Compass, Dr Patrick Cannon from Improbable says “We’re delighted to support Compass, and its ambition to advance the practice of statistics and machine learning, by funding this project. These disciplines are essential to developing the large-scale, high fidelity simulations Improbable’s Defence business is building to give decision makers the information and understanding they need to address complex real-world problems. We’re looking forward to sharing our experience and seeing the new insights this research will generate.”

Compass is a 4-year PhD training programme focusing on Computational Statistics and Data Science. This new venture with Improbable is part of the Compass mission to promote academic and professional agility in its students, equipping them with the skills and experience to work across disciplines in academia and beyond.

The Director of Compass, Prof Nick Whiteley comments “This new partnership between Compass and Improbable brings together world leading expertise to tackle statistical learning problems which arise when trying to explain and predict the behaviour of complex real-world systems. It’s a wonderful opportunity for Compass students to broaden their horizons and develop industry awareness through interaction with Improbable, and an excellent starting point for Dan’s journey as a Data Science researcher.”

Dan Ward is the PhD student recruited to this exciting new project, “I am excited to be starting on the Compass programme, in which I will have the opportunity to work with strong statisticians at both Bristol and Improbable. Agent-based models provide a flexible way to model various real-world systems; however, they can be challenging to calibrate. My project will aim to contribute to statistical methods for calibrating agent-based models, utilising simulations which will run on Improbable’s platform.”

Daniel Ward will be co-supervised by Professor Mark Beaumont (Professor of Statistics in the School of Biological Sciences), Dr Matteo Fasiolo (Lecturer in Statistical Science in the School of Mathematics) and, from Improbable, Dr Patrick Cannon