Short bio

I am currently a Ph.D. student in Machine Learning at Université Laval, supervised by Pascal Germain and François Laviolette. I am a founding member of Baseline, a cooperative of data scientists.

Research Interests

Representation Learning

  • Deep Neural Networks
  • Kernel Methods

Machine Learning Theory

  • PAC-Bayes
  • Sample Compression

Bioinformatic

  • Antibiotic Resistance
  • Drug Discovery

Interpretability

  • Rule-based Models
  • Attention Mechanisms

Publications

Conference Papers

  1. Gaël Letarte, Pascal Germain, Benjamin Guedj, François Laviolette (NeurIPS 2019).
    Dichotomize and Generalize: PAC-Bayesian Binary Activated Deep Neural Networks .
    [ pdf ] • [ link ] • [ GitHub ] • [ poster ] • [ Youtube ]
  2. Gaël Letarte, Emilie Morvant, Pascal Germain (AISTATS 2019).
    Pseudo-Bayesian Learning with Kernel Fourier Transform as Prior .
    [ pdf ] • [ link ] • [ GitHub ] • [ poster ]

Journal Papers

  1. Alexandre Drouin, Gaël Letarte, Frédéric Raymond, Mario Marchand, Jacques Corbeil, François Laviolette (Scientific Report 2019).
    Interpretable genotype-to-phenotype classifiers with performance guarantees.
    [ pdf ] • [ supp ] • [ link ] • [ GitHub ]

Workshops

  1. Gaël Letarte, Frédérik Paradis, Philippe Giguère, François Laviolette (EMNLP 2018).
    Importance of Self-Attention for Sentiment Analysis.
    [ pdf ] • [ link ] • [ GitHub ] • [ poster ]
  2. Alexandre Drouin, Frédéric Raymond, Gaël Letarte, Mario Marchand, Jacques Corbeil, François Laviolette (NIPS 2016).
    Large scale modeling of antimicrobial resistance with interpretable classifiers.
    [ pdf ] • [ link ] • [ GitHub ]

Projects

Tool allowing to learn interpretable computational phenotyping models from k-merized genomic data.

Cooperative of data scientists operating in Quebec City and striving to democratize artificial intelligence.

A Keras-like framework for PyTorch and handles much of the boilerplating code needed to train neural networks.

Academic Affiliations

Want to know more?