This event is part of Decision Making and Uncertainty View Details

Machine Learning and Mean-Field Games

May 23 — 27, 2022

Description

Back to top

Multi-agent reinforcement learning (MARL) with incorporation of techniques and ideas from the theory of mean field games is one of the most active areas in learning and control. The purpose of this workshop is to bring leading experts and junior researchers to  showcase the latest developments in this interdisciplinary field.

Organizer

Back to top
X G
Xin Guo University of California, Berkeley

Speakers

Back to top
B A
Beatrice Acciaio ETH Zürich
H C
Haoyang Cao The Alan Turing Institute
R C
Rene Carmona Princeton University
R C
Rama Cont University of Oxford
C C
Christa Cuchiero University of Vienna
J F
Jean-Pierre Fouque University of California, Santa Barbara
C G
Camilo Garcia University College London
O G
Oliver Gueant Université Paris 1 Pantheon Sorbonne
A H
Anran Hu University of California, Berkeley
R H
Ruimeng Hu University of California, Santa Barbara
S J
Sebastian Jaimungal University of Toronto
D L
Daniel Lacker Columbia University
M L
Martin Larsson Carnegie-Mellon University
M L
Mathieu Lauriere Google Brain
C L
Charles Lehalle Abu Dhabi Investment Authority
S L
Siting Liu University of California, Los Angeles
T M
Thibaut Mastrolia University of California, Berkeley
S P
Sarah Perrin Université de Lille
H P
Huyen Pham University of Paris 6 and CNRS
Z R
Zhenjie Ren University Paris Dauphine – PSL
M R
Mathieu Rosenbaum Ecole Polytechnique
J R
Johannes Ruf London School of Economics
L S
Lukas Szpruch University of Edinburgh
W T
Wenpin Tang Columbia University
J T
Josef Teichmann ETH Zürich
R X
Renyuan Xu University of Southern California
Y Z
Yufei Zhang London School of Economics and Political Science
X Z
Xunyu Zhou Columbia University

Schedule

Back to top
Monday, May 23, 2022
8:30-9:00 CDT
Policy Gradient is Essentially Policy Evaluation

Speaker: Xunyu Zhou (Columbia University)

9:10-9:40 CDT
Optimal Stopping via Randomized Neural Networks

Speaker: Josef Teichmann (ETH Zürich)

9:40-10:30 CDT
Coffee Break
10:30-11:10 CDT
Market making and incentives design in the presence of a dark pool: a deep reinforcement learning approach

Speaker: Mathieu Rosenbaum (Ecole Polytechnique)

11:20-12:10 CDT
Policy gradients algorithms for mean-field control in continuous time

Speaker: Huyên Pham (University of Paris 6 and CNRS)

12:15-13:30 CDT
Lunch
13:30-14:10 CDT
From stochastic control to continuous-time reinforcement learning

Speaker: Yufei Zhang (London School of Economics and Political Science)

14:15-14:55 CDT
Convergence of Empirical Measures, Mean-Field Games and Deep Learning Algorithms

Speaker: Ruimeng Hu (University of California, Santa Barbara)

15:00-15:15 CDT
Coffee Break
15:20-16:00 CDT
Dynamics of Market Making Algorithms in Dealer Markets: Learning and Tacit Collusion

Speaker: Rama Cont (University of Oxford)

Tuesday, May 24, 2022
8:30-9:00 CDT
Deep Learning for Principal-Agent Mean Field Games

Speaker: Sebastian Jaimungal (University of Toronto)

9:10-9:40 CDT
Agency problem with mean field agents

Speaker: Thibaut Mastrolia (University of California, Berkeley (UC Berkeley))

9:40-10:30 CDT
Coffee Break
10:30-11:10 CDT
Propagation of chaos for maxima of particle systems with mean-field drift interaction

Speaker: Martin Larsson (Carnegie-Mellon University)

11:20-12:10 CDT
Minimum curvature flow and martingale exit times

Speaker: Johannes Ruf (London School of Economics)

12:15-13:30 CDT
Lunch
13:30-14:10 CDT
System Noise and Individual Exploration in Learning Large Population Games

Speaker: Renyuan Xu (University of Southern California)

14:15-14:55 CDT
DISTRIBUTIONALLY ROBUST LEARNING OVER DEEP NEURAL NETWORKS AND THEIR ASSOCIATED REGULARIZED RISK

Speaker: Camilo A Garcia Trillos (University College London)

15:00-15:15 CDT
Coffee Break
15:20-16:00 CDT
Some stories about Brownian interacting systems with absorption

Speaker: Wenpin Tang (Columbia University)

Wednesday, May 25, 2022
8:30-9:00 CDT
Market making algorithms: the next success of Reinforcement Learning?

Speaker: Olivier Gueant (Université Paris 1 Pantheon Sorbonne)

9:10-9:40 CDT
Optimal bailout strategies resulting from the drift controlled supercooled Stefan problem

Speaker: Christa Cuchiero (University of Vienna)

9:40-10:30 CDT
Coffee Break
10:30-11:10 CDT
Exploration-exploitation trade-off for continuous-time episodic reinforcement learning

Speaker: Lukas Szpruch (University of Edinburgh)

11:20-12:10 CDT
Mean Field Optimization regularized by Fisher Information

Speaker: Zhenjie Ren (University Paris Dauphine – PSL)

12:15-13:30 CDT
Lunch
13:30-13:55 CDT
Scalable Deep Reinforcement Learning Algorithms for Mean Field Games

Speaker: Sarah Perrin (Université de Lille)

14:00-14:25 CDT
An Optimization Framework for Solving Mean-Field Games

Speaker: Anran Hu (University of California, Berkeley (UC Berkeley))

14:25-14:35 CDT
Coffee Break
14:35-15:00 CDT
Controlling the propagation of epidemics via mean-field control

Speaker: Siting Liu (University of California, Los Angeles (UCLA))

15:05-15:30 CDT
Sensitivity and Robustness of Stackelberg Mean-Field Games through an Optimization Lense

Speaker: Jiacheng Zhang (University of California, Berkeley (UC Berkeley))

Thursday, May 26, 2022
8:30-9:00 CDT
Mean field approximations via log-concavity, and a non-asymptotic perspective on mean field control

Speaker: Daniel Lacker (Columbia University)

9:10-9:40 CDT
Sequential learning via adapted transport

Speaker: Beatrice Acciaio (ETH Zürich)

9:40-10:30 CDT
Coffee Break
10:30-11:00 CDT
Reinforcement Learning Algorithm for Mixed Mean Field Control Games

Speaker: Jean-Pierre Fouque (University of California, Santa Barbara (UCSB))

11:10-11:40 CDT
Assessing Transfer Learning

Speaker: Haoyang Cao (The Alan Turing Institute)

11:45-13:30 CDT
Lunch
13:30-14:10 CDT
An Overview of Learning Mean Field Games

Speaker: Mathieu Lauriere (NYU Shanghai)

14:15-14:55 CDT
Model-Free Mean-Field Reinforcement Learning: Mean-Field MDP and Mean-Field Q-Learning

Speaker: Rene Carmona (Princeton University)

15:00-16:00 CDT
Social Hour

Videos

Back to top

Policy Gradient is Essentially Policy Evaluation

Xunyu Zhou
May 23, 2022

Optimal Stopping via Randomized Neural Networks

Josef Teichmann
May 23, 2022

Market making and incentives design in the presence of a dark pool: a deep reinforcement learning approach

Mathieu Rosenbaum
May 23, 2022

Policy gradients algorithms for mean-field control in continuous time

Huyên Pham
May 23, 2022

From stochastic control to continuous-time reinforcement learning

Yufei Zhang
May 23, 2022

Convergence of Empirical Measures, Mean-Field Games and Deep Learning Algorithms

Ruimeng Hu
May 23, 2022

Dynamics of Market Making Algorithms in Dealer Markets: Learning and Tacit Collusion

Rama Cont
May 23, 2022

Deep Learning for Principal-Agent Mean Field Games

Sebastian Jaimungal
May 24, 2022

Agency problem with mean field agents

Thibaut Mastrolia
May 24, 2022

Propagation of chaos for maxima of particle systems with mean-field drift interaction

Martin Larsson
May 24, 2022

Minimum curvature flow and martingale exit times

Johannes Ruf
May 24, 2022

System Noise and Individual Exploration in Learning Large Population Games

Renyuan Xu
May 24, 2022

DISTRIBUTIONALLY ROBUST LEARNING OVER DEEP NEURAL NETWORKS AND THEIR ASSOCIATED REGULARIZED RISK

Camilo A Garcia Trillos
May 24, 2022

Some stories about Brownian interacting systems with absorption

Wenpin Tang
May 24, 2022

Market making algorithms: the next success of Reinforcement Learning?

Olivier Gueant
May 25, 2022

Optimal bailout strategies resulting from the drift controlled supercooled Stefan problem

Christa Cuchiero
May 25, 2022

Exploration-exploitation trade-off for continuous-time episodic reinforcement learning

Lukas Szpruch
May 25, 2022

Mean Field Optimization regularized by Fisher Information

Zhenjie Ren
May 25, 2022

Scalable Deep Reinforcement Learning Algorithms for Mean Field Games

Sarah Perrin
May 25, 2022

An Optimization Framework for Solving Mean-Field Games

Anran Hu
May 25, 2022

Controlling the propagation of epidemics via mean-field control

Siting Liu
May 25, 2022

Sensitivity and Robustness of Stackelberg Mean-Field Games through an Optimization Lense

Jiacheng Zhang
May 25, 2022

Mean field approximations via log-concavity, and a non-asymptotic perspective on mean field control

Daniel Lacker
May 26, 2022

Reinforcement Learning Algorithm for Mixed Mean Field Control Games

Jean-Pierre Fouque
May 26, 2022

Assessing Transfer Learning

Haoyang Cao
May 26, 2022

An Overview of Learning Mean Field Games

Mathieu Lauriere
May 26, 2022

Model-Free Mean-Field Reinforcement Learning: Mean-Field MDP and Mean-Field Q-Learning

Rene Carmona
May 26, 2022