The Canada Excellence Research Chair in Data Science for Real-Time Decision-Making aims at developing new tools and methodologies that will allow enormous volumes of data from multiple sources to be processed and analyzed in real time — in order to obtain usable knowledge and to automate decision-making. By combining processes for analyzing highly targeted data and real-time decision-making, their mathematical model-based tools will help organizations improve performance, by creating highly customized outputs and taking into account the environments, needs and individual behaviors of their clients or users. The applications that result will foster new business models that are based on accurate depictions of user behaviors and expectations, combined with competitors’ responses. The many sectors that could benefit include transportation management, energy, health care and manufacturing, as well as supply chain management and logistics.
The Chair will make it possible to train the next generation of data science specialists, who will be able to master the scientific, technological and economic issues emerging from the big-data explosion. Their multidisciplinary skills and their understanding of business issues will make them a labor force highly sought-after by employers from every sector of the economy who wish to transform their decision-making processes.
The research activities within the Chair has been consolidated through the development of Ecole standing for Extensible Combinatorial Optimization Learning Environments and aiming to expose a number of control problems arising in combinatorial optimization solvers as Markov Decision Processes. Rather than trying to predict solutions to combinatorial optimization problems directly, the philosophy behind Ecole is to work in cooperation with the state-of-the-art Mixed Integer Linear Programming solver SCIP that acts as a controllable algorithm.
[1/3] The Machine Learning for Combinatorial Optimization (ML4CO) NeurIPS 2021 competition aims at improving state-of-the-art combinatorial optimization solvers by replacing / integrating key heuristic components with machine learning models.
Vidéo: le professeur Andrea Lodi présente les activités de la Chaire d'excellence en recherche du Canada sur la science des données pour la prise de décision en temps réel dont il est le titulaire à PolyMTL (en anglais, avec sous-titres en français) https://www.youtube.com/watch?v=CVhGP5sClHI (1/2)
Want to get a high-level overview of the Weisfeiler-Leman algorithm's use in ML and its connection to GNNs? Check out our IJCAI survey track paper: https://arxiv.org/abs/2105.05911.
Joint work with @rusty1s (@sfb876) and Nils M. Kriege (@univienna).
October 2, 2020: Mathieu Tanneau‘s defends his thesis on Exploiting structure in Mixed-Integer Linear and Non-Linear Programming
October 2, 2020: Members of the Chair, Elias Khalil and Aleksandr Kazachkov are hosting a session of Discrete Optimization Talks, featuring Andrea Lodi and Tuomas Sandholm. Details are up on the website: https://talks.discreteopt.com
October 1, 8, 15, 22, 29 2020: This first-ever edition of IVADO Digital October will showcase student-led research in our IVADO community. All month long, digital intelligence will be front and centre, starting with a distinguished panel of experts on October 1 including Prof. Andrea Lodi and followed by presentations of multidisciplinary projects by our scholarship recipients.
The Chair relies on various public and private partners in hospitals and industrial sectors to play an active role in technological, economic and social development.