Deep Learning across games by Daniele Condorelli

Join us for another event in the Microeconomics Research Seminar Series, Autumn Term 2025.

Daniele Condorelli, from the University of Warwick, will present this week's Microeconomics Seminar on Deep Learning across games.

Abstract

We train two neural networks adversarially to play static games. At each iteration, a row and column network observe a new random bimatrix game and output individual mixed strategies. The parameters of each network are independently updated via stochastic gradient descent on a loss defined by the individual squared regret experienced in the game. Simulations show the joint behavior of the trained networks approximates a Nash equilibrium in all games. In 2×2 games with multiple equilibria, the networks select the risk dominant equilibrium. These findings, which are robust and generalise out-of-distribution, illustrate how equilibrium emerges from learning across heterogeneous games.

This seminar will be held in the Economics Common Room on Monday 3 November at 4.00pm. This event is open to all levels of study and is also open to the public.

To register your place and gain access to the webinar, please contact the seminar organisers.

This event is part of the Microeconomics Research Seminar Series.