Avatar

Alexandre Drouin

Staff Research Scientist

ServiceNow

Biography

Welcome to my page!

I’m a Staff Research Scientist at ServiceNow Research in Montreal and an Adjunct Professor of Computer Science at Laval University and Mila. I lead ServiceNow’s Human Decision Support research program, which explores the use of machine learning for decision-making in complex dynamic environments.

My main research interest is causal decision-making under uncertainty, where the goal is to answer questions of causal nature (interventions, counterfactual), while accounting for sources of uncertainty, such as ambiguity in causal structures and unmeasured variables. I am also interested in probabilistic time series forecasting and its use to foresee the long-term effect of actions.

In the past, I developed machine learning algorithms for biomarker discovery in large genomic datasets and focussed on applications to the global problem of antibiotic resistance. I obtained my Ph.D. in 2019 under the supervision of François Laviolette.

Interests

  • Machine learning
  • Causal inference and discovery
  • Time series forecasting
  • Model robustness

Education

  • PhD in Artificial Intelligence, 2019

    Laval University

  • MSc in Artificial Intelligence, 2014

    Laval University

  • BSc in Computer Science, 2012

    Laval University

Publications

Quickly discover relevant content by filtering publications.
(2023). Causal Discovery with Language Models as Imperfect Experts. ICML 2023 Workshop on Structured Probabilistic Inference & Generative Modeling.

PDF Code

(2023). Invariant Causal Set Covering Machines. ICML 2023 Workshop on Spurious Correlations, Invariance, and Stability.

PDF Code

(2023). Regions of Reliability in the Evaluation of Multivariate Probabilistic Forecasts. 40th International Conference on Machine Learning (ICML 2023).

PDF Code

(2022). TACTiS: Transformer-Attentional Copulas for Time Series. 39th International Conference on Machine Learning (ICML 2022) – Spotlight Presentation.

PDF Code

(2022). Typing Assumptions Improve Identification in Causal Discovery. 1st Conference on Causal Learning and Reasoning (CLeaR 2022) – Oral Presentation.

PDF Code

Talks

Multivariate probabilistic time series forecasting: transformer-based copulas and the limitations of proper scoring rules

Accurate estimation of time-varying quantities is crucial for effective decision-making across various domains, including healthcare …

Apprentissage automatique et raisonnement causal - Au delà des corrélations

La recherche en apprentissage automatique a rendu possibles d’époustouflantes avancées en intelligence artificielle. Un avenir où un …

Mind the structure: an introduction to causal inference for machine learning

Machine learning algorithms excel at discovering statistical dependency patterns in data. Those can be exploited to produce extremely …

Tutorial: A practical introduction to causal inference

This tutorial will consist of a practical introduction to the estimation of causal effects. We will experiment with the concepts of …

Differentiable Causal Discovery with Observational and Interventional Data

Knowledge of the causal structure that underlies a data generating process is essential to answering questions of causal nature. Such …

Current students

Graduate Students

Avatar

Arjun Ashok

PhD Student

Avatar

Thibaud Godon

PhD Student