Nicholas Bishop 'Machine Learning and ABMs...'

Who Am I?

I am a postodctoral researcher at the University of Oxford. More specifically, I work within the Department of Computer Science on the Large Agent Collider (LAC) project. During my PhD at the Univeristy of Southampton, I studied a variety of problems at the intersection of mechanism design and machine learning under the joint supervision of Prof. Long Tran-Thanh and Prof. Enrico Gerding.

I am currently focused on applying state-of-the-art machine learning methods to rapidly calibrate, analyse and validate agent-based models (ABMs) of large-scale complex systems. My recent work includes the use of generalised variational inference to search for realistic synthetic populations, and finding interpretable surrogate models for ABMs through causal abstraction.

What is the LAC Project?

The overarching technical goal of the LAC project is to effect a step change in our ability to develop and deploy robust large scale agent-based simulations by leveraging techniques from AI and machine learning. The LAC project is supervised jointly by Prof. Michael Wooldridge, Prof. Ani Calinescu, and Prof. Doyne Farmer.

Contacting Me

If you are an agent-based modeller interested in how machine learning might be applied to your problem, or conversely a machine learning specialist interested in dicussing the unique challenges ABMs present, feel free to shoot me an email.

Email: nicholas.bishop@cs.ox.ac.uk

Other Stuff

I maintain a blog post on this website which aims to provide updates about recently accepted papers. I also occassionally write about general ideas in machine learning that pique my interest. Additionally, I am fairly active on Twitter.