AI4ABM - International AI for Agent-Based Modelling Community

Agent-Based Modelling (ABM) has seen increasing interest across various disciplines, ranging from economics and epidemiology to cybersecurity, social sciences, and climate policy. At the same time, recent progress in AI has yielded exciting new approaches to learning, calibrating, and analysing such models, including deep multi-agent learning and simulation-based inference. This interdisciplinary community is meant to bring together practitioners and theorists to boost ABM method development in AI, and stimulate novel applications across disciplinary boundaries.

New to the space of AI and/or Agent-Based Modeling? Take a look at our tutorial pages.

Click here for information on our hybrid-remote biweekly reading group based at the University of Oxford UK

Please follow us on Twitter:

news

May 25, 2022 We now have a twitter account - please follow
Apr 21, 2022 Prof. Neil Ferguson (Imperial College London) is joining us remotely to talk about COVID-19 and ABMs. Location TBA.
Apr 7, 2022 The Large Agent Collider project kickoff meeting is taking place in St Hilda’s College, Oxford (it runs until April 8th).
Mar 18, 2022 We are stoked to announce that our workshop proposal “AI for Agent-Based Modelling” was accepted at the International Conference on Machine Learning taking place in Baltimore, Maryland from July 19th-24th. We have a number of fantastic speakers lined up and will soon publish our Call for Papers on this website.
Mar 10, 2022 We are excited to host Sören Mindermann, who will talk about his work on modelling COVID-19 interventions.
Feb 24, 2022 This will be the second session of the AI4ABM reading group. Note our changed time: 4:10pm at LR7! We will be discussing this paper.
Feb 10, 2022 The AI4ABM Reading Group launched successfully today, drawing a diverse crowd of 25+ AI and ABM practitioners, both in person and remote. We are excited to have reached critical mass already in the first meeting, and are looking forward to getting work done!