Events at The University of Manchester
  • University home
  • Events
  • Home
  • Exhibitions
  • Conferences
  • Lectures and seminars
  • Performances
  • Events for prospective students
  • Sustainability events
  • Family events
  • All Events

AI-Fun Seminar | Robert Loftin: Learning, Negotiation and Cooperation in Differentiable Games

image
Dates:14 February 2024
Times:10:30 - 11:30
What is it:Seminar
Organiser:Faculty of Science and Engineering
Who is it for:University staff, External researchers, Current University students
Speaker:Dr Robert Loftin
See travel and contact information
Add to your calendar

Other events

  • In category "Seminar"
  • In group "(DF) Data Science and AI"
  • By Faculty of Science and Engineering

The Manchester Centre for AI Fundamentals is hosting a series of seminars featuring expert researchers working in the fundamentals of AI.

On 14 February, Dr Robert Loftin joins us from The University of Sheffield with a talk titled 'Learning, Negotiation and Cooperation in Differentiable Games'.

Abstract:

In tasks that require a high degree of cooperation between independent agents, success often on the ability of these agents to adapt to one another’s behavior. Simultaneous adaptation, however, can prevent agents from converging to a fixed joint strategy at all, let alone an optimal strategy. To address this issue, we can model cooperative learning as a Stackelberg competition, in which the “leader” agent selects the strategy that maximizes its own payoff given that the “follower” agent will choose their best response to this strategy. In this talk I will discuss recent work on the problem of "uncoupled" learning of Stackelberg solutions in differentiable games. In the uncoupled setting, individual learners are only aware of their own utility functions, and have limited information about their partner's learning strategy. This requires the leader to estimate the follower's best-response function using only their observable behavior. It also requires a mechanism by which agents can "negotiate" who will lead and who will follow. This talk will present a new learning algorithm that addresses both of these issues. I will also discuss the value of differentiable games as a model of long-term human-AI cooperation, and the implications of theoretical results in such games for the larger problem of building cooperative AI.

Speaker

Dr Robert Loftin

Role: Lecturer in Machine Learning

Organisation: University of Sheffield

  • https://www.sheffield.ac.uk/dcs/people/academic/robert-loftin

Travel and Contact Information

Find event

Emmeline Suite
Christabel Pankhurst Building
Dover Street
Manchester

Contact event

Centre for AI Fundamentals

ai-fun@manchester.ac.uk

Contact us

  • +44 (0) 161 306 6000

Find us

The University of Manchester
Oxford Rd
Manchester
M13 9PL
UK

Connect with the University

  • Facebook page for The University of Manchester
  • X (formerly Twitter) page for The University of Manchester
  • YouTube page for The University of Manchester
  • Instagram page for The University of Manchester
  • TikTok page for The University of Manchester
  • LinkedIn page for The University of Manchester

  • Privacy /
  • Copyright notice /
  • Accessibility /
  • Freedom of information /
  • Charitable status /
  • Royal Charter Number: RC000797
  • Close menu
  • Home
    • Featured events
    • Today's events
    • The Whitworth events
    • Manchester Museum events
    • Jodrell Bank Discovery Centre events
    • Martin Harris Centre events
    • The John Rylands Library events
    • Exhibitions
    • Conferences
    • Lectures and seminars
    • Performances
    • Events for prospective students
    • Sustainability events
    • Family events
    • All events