ml-udm.github.io

rue d'Ulm

Workshop on Machine Learning and User Decision Making

Dates: May 23-24, 2019

Venue: 29 rue d’Ulm, Paris, France (special note: access to registered attendees, not affiliated with ENS, will be through 24, rue Lhomond)

Organizers: Laurent Massoulie (MSR-INRIA Joint Centre) and Milan Vojnovic (London School of Economics)

Scope: The scope of this workshop is to discuss important algorithmic and mathematical questions that arise at the interface between machine learning and user decision making. The topics include but are not limited to online matching in uncertain environments, reinforcement learning, optimization and learning, mechanism design for online platforms, and recommender systems.

Please contact the organizers if you would like to attend.

Tentative program schedule

Time Day 1 - Thursday May 23, 2019
08:45-
09:00
Welcome note
09:00-
09:55
Online Algorithms, Learning and Game Theory: Linear, Convex and Beyond
Nikhil Devanur, Microsoft Research
Abstract
10:00-
10:55
Exploration in Structured Reinforcement Learning
Alexandre Proutière, KTH
Abstract
11:00-
11:30
Coffee break
11:30-
12:25
Can RL Learn Classic Optimization Algorithms
Aranyak Mehta, Google Research
Abstract
12:30-
14:00
Lunch break
14:00-
14:55
Optimal Control in Dynamic Matching Systems
Ana Busic, INRIA
Abstract
15:00-
15:55
State Dependent Control of Ride-Hailing Systems
Yash Kanoria, Columbia University
Abstract
16:00-
16:55
On The Limits and Potential Solutions for state-of-the-art Recommendation
Flavian Vasile, Criteo
Abstract



 
Time Day 2 - Friday May 24, 2019
09:00-
09:55
On Reinforcement Learning using Monte Carlo Tree Search with Supervised Learning: Non-Asymptotic Analysis
Devavrat Shah, MIT
Abstract
10:00-
10:55
Hierarchical Clustering
Claire Mathieu, CNRS
Abstract
11:00-
11:30
Coffee break
11:30-
12:25
Optimal Auctions through Deep Learning
Paul Duetting, LSE
Abstract
12:30-
14:00
Lunch break
14:00-
14:55
Adaptive Experimental Design with Temporal Interference
Ramesh Johari, Stanford University
Abstract
15:00-
15:55
Econometrics and Machine Learning through the Lens of Neyman Orthogonality
Vasilis Syrgkanis, Microsoft Research
Abstract
16:00-
16:55
Decision Theoretic Approach for Improved Training of Generative Adversarial Networks
Sewoong Oh, University of Washington
Abstract