----------------------------------------------------------------------------------------------------

          Department of Systems Engineering and Engineering Management

                                  The Chinese University of Hong Kong

----------------------------------------------------------------------------------------------------

Date: Monday, June 15, 2026, 4:30pm to 5:30pm HKT

Venue: ERB513, The Chinese University of Hong Kong

Title: Reinforced-GANs for Market Equilibrium Models

Speaker: Professor Xiaofei Shi, University of Toronto

Abstract:

We present a general numerical framework for solving continuous-time financial market equilibria under minimal modeling assumptions while incorporating realistic financial frictions, such as trading costs, and supporting multiple interacting agents. Inspired by generative adversarial networks (GANs) and classical Expectation-Maximization (EM) algorithms, our approach employs a generative deep reinforcement learning framework with a decoupling feedback system embedded in the adversarial training loop, which we term the reinforcement link. This architecture, which we name Reinforced-GAN, stabilizes the training dynamics by explicitly incorporating learned feedback from the discriminator. Our theoretically guided feedback mechanism leads to a decoupled treatment whose analysis does not rely on the small time horizon assumptions typically required for fully coupled forward backward stochastic differential equation (FBSDE) systems, at the level of the generator and discriminator sub‑routines, thereby mitigating challenges that hinder conventional numerical algorithms. Experimentally, our algorithm not only learns but also provides testable predictions on how asset returns and volatility emerge from the endogenous trading behavior of market participants, where traditional analytical methods fall short. 

Biography:

Xiaofei Shi is an Assistant Professor in the Department of Statistical Sciences at the University of Toronto. Before joining University of Toronto, she worked as a Term Assistant Professor at Columbia University and obtained her PhD in Mathematical Finance at Carnegie Mellon University. Her research interests lie in stochastic control and stochastic optimization, particularly their applications in portfolio management, asset pricing, and general equilibrium modeling.

Everyone is welcome to attend the talk!

SEEM-5202 Website: http://seminar.se.cuhk.edu.hk