Introduction

Multi-agent systems (MAS) are a core area of research of contemporary artificial intelligence. A multi-agent system consists of multiple decision-making agents which interact in a shared environment to achieve common or conflicting goals. MAS research spans a range of technical problems, such as how to design MAS to incentivise certain behaviours in agents, how to design algorithms enabling one or more agents to achieve specified goals in a MAS, how information is communicated and propagated among agents, and how norms, conventions and roles may emerge in MAS. A vast array of applications can be addressed using MAS methodologies, including autonomous driving, multi-robot factories, automated trading, commercial games, automated tutoring, etc.

The MAS interest group at The Alan Turing Institute provides a central meeting point for UK-based researchers in universities, industry, and defence who are interested in MAS research. We maintain a UK-wide MAS mailing list through which list members can communicate relevant opportunities and events. We organise regular events (spotlight presentations, UK-MAS symposia) and maintain a virtual MAS labs map shown below.

 

 

Explaining the science

Watch interest group organisers Stefano Albrecht and Michael Wooldridge discuss MAS work in more detail in the below videos.

Aims

The principle activities of the MAS group are the following:

  • Maintain a UK-wide MAS mailing list to communicate relevant events and opportunities at the Turing and beyond.

  • Organise a monthly spotlight virtual meeting in which selected speakers (university lab PIs and PhD students, industry labs) have an opportunity to present their recent MAS projects.

  • Maintain the UK-MAS virtual labs map to understand the landscape of current MAS research in the UK at universities and industry.

Multi-Agent Systems Talk Series

The group organises a Multi-Agent Systems Talk Series which takes place once a month to give UK-based labs an opportunity to present their MAS research and initiatives. Sign up to our mailing list to receive updates about upcoming talks. If you would like to give a talk, please contact the organisers.

Date Speaker Title Link to watch
24 February 2021, 10:00 Allan Dafoe
Edward Hughes
Open Problems in Cooperative AI Watch now
26 March, 13:00 Kobi Gal Behaviour Change for Social Good using AI Watch now
22 April, 11:00 Nick Hawes
Bruno Lacerda
Towards Guaranteed Multi-Robot Systems Under Uncertainty Watch now
20 May, 10:00 Subramanian Ramamoorthy
Helen Hastie

Trustworthy Autonomous Systems

Watch now
24 June, 10:00 Long Tran-Thanh Optimisation Problems in Human-Agent Learning Watch now
15 July, 14:00 Sam Devlin

Coordinated Self-Play to Ad-Hoc Teamwork in Bleeding Edge

Watch now
26 August, 10:00 Shimon Whiteson
Tabish Rashid
Factored Value Functions for Cooperative Multi-Agent Reinforcement Learning Watch now
23 September, 14:00 Alison Heppenstall
Nick Malleson 
TBC Coming soon
21 October, 14:00 Alessio Lomuscio TBC Coming soon

 

Talking points

Is human intelligence inherently social?

Will AI require analogues of human social skills, such as cooperation, coordination, and negotiation?

Should machines use human languages to communicate with each other?

How do norms and languages emerge?

How can an autonomous agent interact safely and reliably with other agents and humans to achieve its goals?

How can research in multi-agent systems effectively translate into industry?

How to get involved

  1. Sign up to our moderated multi-agent systems mailing list to receive updates about relevant opportunities
  2. Join the members group, where you can have discussions with other members, share research, events and opportunities

Join the mailing list and members group

Organisers