Hi!

I'm a PhD student in computer science at Université de Montréal/Mila. I am broadly interested in generative modeling and topics bridging computer science, economics and mathematics.

Papers


Expected Flow Networks in Stochastic Environments and Two-Player Zero-Sum Games

Marco Jiralerspong*, Bilun Sun*, Danilo Vucetic*, Tianyu Zhang, Yoshua Bengio, Gauthier Gidel, Nikolay Malkin

ICLR 2024.

An extension of GFlowNets to stochastic environments (EFlowNets) and adversarial environments (AFlowNets) with satisfiable constraints.

On the Stability of Iterative Retraining of Generative Models on their own Data

Quentin Bertrand, Joey Bose, Alexandre Duplessis, Marco Jiralerspong, Gauthier Gidel

ICLR 2024 (spotlight).

A theoretical and experimental analysis of stability of generative models when they are trained with their own generated samples.

Feature Likelihood Divergence: Evaluating the Generalization of Generative Models Using Samples

Marco Jiralerspong, Joey Bose, Ian Gemp, Chongli Qin, Yoram Bachrach, Gauthier Gidel

NeurIPS 2023.

A novel likelihood-based metric for evaluating generative models which accounts for overfitting (formally known as the Feature Likelihood Score, FLS).

Generating Diverse Vocal Bursts with StyleGAN2 and MEL-Spectrograms

Marco Jiralerspong, Gauthier Gidel

Expressive Vocalisation Workshop & Competition, ICML 2022.

Training on StyleGAN2 on MEL-Spectrograms of vocal bursts to generate diverse samples corresponding to various emotions.

Regret Dynamics in Online Clustering

Marco Jiralerspong*, Andjela Mladenovic*

IFT 6269, Fall 2021

Analysis of regret dynamics of an online Follow-The-Leader (FTL) k-means algorithm and an online FTL Gaussian Mixture Model (GMM).

Comparison of Losses Used by Generative Adversarial Networks

Marco Jiralerspong

COMP 598, Fall 2019

Course project for COMP 598 (Mathematical Foundations for Machine Learning) comparing various losses used for the training of GANs.

Projects


Gale-Shapley Interactive Simulation | Interactive Simulation

An interactive, step-by-step, visualization of the Gale-Shapley algorithm in action for random preferences.

Python for Biologists | Github

An introductory Python course for biologists explaining how to get set up, the basic elements of the language and progressively harder exercises.

DPYD Analysis | Github

Data analysis of DPYD variants demonstrating the prevalence of LOF variants in minority populations not considered for testing by provincial guidelines.

Marischedule | Github

Mock schedule builder.

Notes


LaTeX notes (except for one, from the dark days before I learned about LaTeX) for various courses at McGill University: