Portrait de Erick Delage

Erick Delage

Membre académique associé
erick.delage@hec.ca
Professeur titulaire, HEC Montréal, Département de sciences de la décision
Sujets de recherche
Apprentissage par renforcement
Optimisation
Site web
Google Scholar
LinkedIn

Biographie

Erick Delage est professeur au Département de sciences de la décision à HEC Montréal, titulaire de la Chaire de recherche du Canada en prise de décision sous incertitude, et membre du Collège de nouveaux chercheurs et créateurs en art et en science de la Société royale du Canada. Ses domaines de recherche englobent l'optimisation robuste et stochastique, l'analyse de décision, l'apprentissage automatique, l'apprentissage par renforcement et la gestion des risques avec des applications en optimisation de portefeuille, en gestion des stocks, et dans les problèmes liés à l'énergie et aux transports.

Étudiants actuels

Nima Akbarzadeh
Postdoctorat - HEC
Abhilash Chenreddy
Doctorat - HEC
Esther Derman
Postdoctorat - UdeM
Superviseur⋅e principal⋅e :
Pierre-Luc Bacon
Site web
Github
Google Scholar
Xiaohui Tu
Doctorat - HEC
Weikai Wang
Doctorat - HEC
Xin Wang
Doctorat - HEC

Publications

Robust Data-driven Prescriptiveness Optimization
Mehran Poursoltani
Erick Delage
Angelos Georghiou
The abundance of data has led to the emergence of a variety of optimization techniques that attempt to leverage available side information t… (voir plus)o provide more anticipative decisions. The wide range of methods and contexts of application have motivated the design of a universal unitless measure of performance known as the coefficient of prescriptiveness. This coefficient was designed to quantify both the quality of contextual decisions compared to a reference one and the prescriptive power of side information. To identify policies that maximize the former in a data-driven context, this paper introduces a distributionally robust contextual optimization model where the coefficient of prescriptiveness substitutes for the classical empirical risk minimization objective. We present a bisection algorithm to solve this model, which relies on solving a series of linear programs when the distributional ambiguity set has an appropriate nested form and polyhedral structure. Studying a contextual shortest path problem, we evaluate the robustness of the resulting policies against alternative methods when the out-of-sample dataset is subject to varying amounts of distribution shift.
2023-06-09
ArXiv (prépublication)
doi.org
arxiv.org
Technical Note—Risk-Averse Regret Minimization in Multistage Stochastic Programs
Mehran Poursoltani
Erick Delage
Angelos Georghiou
2023-01-30
Operational Research (publié)
doi.org
On Dynamic Program Decompositions of Static Risk Measures
Jia Lin Hau
Erick Delage
Mohammad Ghavamzadeh
Marek Petrik
Optimizing static risk-averse objectives in Markov decision processes is challenging because they do not readily admit dynamic programming d… (voir plus)ecompositions. Prior work has proposed to use a dynamic decomposition of risk measures that help to formulate dynamic programs on an augmented state space. This paper shows that several existing decompositions are inherently inexact, contradicting several claims in the literature. In particular, we give examples that show that popular decompositions for CVaR and EVaR risk measures are strict overestimates of the true risk values. However, an exact decomposition is possible for VaR, and we give a simple proof that illustrates the fundamental difference between VaR and CVaR dynamic programming properties.
2023-01-01
arXiv.org (prépublication)
doi.org
Data-Driven Optimization with Distributionally Robust Second Order Stochastic Dominance Constraints
Chun Peng
Erick Delage
This paper presents the first comprehensive study of a data-driven formulation of the distributionally robust second order stochastic domina… (voir plus)nce constrained problem (DRSSDCP) that hinges on using a type-1 Wasserstein ambiguity set. It is, furthermore, for the first time shown to be axiomatically motivated in an environment with distribution ambiguity. We formulate the DRSSDCP as a multistage robust optimization problem and further propose a tractable conservative approximation that exploits finite adaptability and a scenario-based lower bounding problem. We then propose the first exact optimization algorithm for this DRSSDCP. We illustrate how the data-driven DRSSDCP can be applied in practice on resource-allocation problems with both synthetic and real data. Our empirical results show that, with a proper adjustment of the size of the Wasserstein ball, DRSSDCP can reach acceptable out-of-sample feasibility yet still generating strictly better performance than what is achieved by the reference strategy.
2022-11-22
Operational Research (publié)
doi.org