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

    Department of Systems Engineering and Engineering Management

                       The Chinese University of Hong Kong

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

Date: Friday, April 26, 4:30 pm – 5:30 pm

Venue: ERB 513, The Chinese University of Hong Kong

Title: Prediction and Prescription: Some Views on the Interplay Between Machine Learning and Decision-Making Under Uncertainty

Speaker:  Prof. Bernardo Pagnoncelli, SKEMA Business School

Abstract:

In recent years, there has been a notable surge in the number of articles related to end-to-end learning, also known as decision-focused learning. The fundamental concept revolves around devising models that connect raw data to decisions, often by incorporating an intermediary step involving some form of prediction. I will delve into prominent paradigms within the literature that adhere to this data-driven approach, focusing on contextual optimization methods. Additionally, I will present various models from Stochastic Programming and simulation optimization that assume a decision-maker can observe the state of the world before making a decision.
I will show theoretical results that validate those approaches, discuss under which circumstances one method is more appropriate than others, and show some numerical examples that illustrate the efficacies of those approaches.
This is joint work with Hamed Rahimian, Barry Nelson, Gregory Keslin, Matthew Plumlee, Arturo Cifuentes and Domingo Ramírez

Biography:

Bernardo K. Pagnoncelli obtained a Ph.D. in Applied Mathematics from PUC-Rio, Brazil, in 2009, with a one-year internship (2007) at Georgia Tech, United States. He is a Full Professor at SKEMA Business School in Lille, France, and is the Director of the SKEMA-KU Leuven Ph.D. program, in the track of Artificial Intelligence and Operations Management. He conducts research in the area of optimization under uncertainty, with applications to finance, pension funds, energy, and natural resources management. He has publications in various prestigious journals in the field of Operations Research, and his research has been funded by the Chilean funding agency and companies such as Petrobras (Brazil) and EDF (France). He is currently a member of COSP, the committee of the Stochastic Optimization Society, and was co-chair of the Nicholson Prize 2020, the main prize for students awarded by INFORMS. In 2016, he was a visiting professor (3 months) at Texas A&M University, College Station, United States. In 2018-2019, he received the Patrick and Amy McCarter Fellowship, which funded his one-year stay as a visiting Associate Professor at Northwestern University, Evanston, United States.

Everyone is welcome to attend the talk!

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

Email: seem5202@se.cuhk.edu.hk