We are recruiting a PhD student to work on responsible and robust decision-focused learning. The project will develop machine learning models that are trained not only to make accurate predictions, but also to support high-quality downstream decisions in optimization problems.
The position is well suited for a student interested in the intersection of operations research, machine learning, stochastic optimization, bilevel optimization, and game theory. Possible research directions include interpretable decision policies, scalable prediction-and-optimization pipelines, and robust learning methods for settings where data may be noisy, strategically manipulated, or privacy-sensitive.
The student will be part of a collaborative research environment across Universite de Montreal, Polytechnique Montreal, GERAD, CIRRELT, and Mila. The project will also connect to applications in energy, transportation, and supply chains, where reliable decision-support systems are increasingly important.
Strong applicants should have a solid mathematical background and good programming skills. Prior exposure to optimization, machine learning, probability, or algorithms is valuable. A master’s degree is helpful but not strictly required for exceptional candidates.
Related reading:
- A survey of contextual optimization methods for decision-making under uncertainty, by Utsav Sadana, Abhilash Chenreddy, Erick Delage, Alexandre Forel, Emma Frejinger, and Thibaut Vidal.
- Solving two-stage programs with endogenous uncertainty via random variable transformation, by Maria Bazotte, Margarida Carvalho, and Thibaut Vidal.
- Trained random forests completely reveal your dataset, by Julien Ferry, Ricardo Fukasawa, Timothee Pascal, and Thibaut Vidal.
Interested candidates should email Thibaut Vidal, Margarida Carvalho, and Utsav Sadana, with the subject line PhD application - Responsible Decision-Focused Learning and include:
- a CV;
- transcripts, if available;
- a short statement describing research interests and preparation;
- links to papers, code, or projects, if available;
- contact information for references.
Applications will be reviewed on a rolling basis. We are committed to an inclusive research environment and encourage applications from candidates of all backgrounds, including members of groups that are underrepresented in mathematics, computer science, operations research, and engineering.