I am pursuing a Ph.D. degree at Mila and the Université de Montréal, supervised by Professor Simon Lacoste-Julien. I also worked for two years at Meta Fundamental Artificial Intelligence Research (FAIR) in Montreal as a visiting researcher, supervised by Professor Pascal Vincent and collaborating with Dr. Kartik Ahuja. I spent the past summer in New York as a NeuroAI intern at the Cold Spring Harbor Laboratory, working with Professor David Klindt.

I am currently on the job market for research positions. Please reach out if you see a fit.

I have been conducting research on causal representation learning, independent component analysis, and disentanglement, and out-of-distribution generalization. I’m also broadly interested robustness, safety, interpretability, fairness, and AI for science.

I have a master’s degree in data science from the University of Helsinki, where I was supervised by Professor Aapo Hyvärinen and Dr. Antti Hyttinen, working on independent component analysis for binary data (thesis).

I grew up in Brazil and moved to Finland to do my bachelor’s in computer science at the University of Helsinki. During my bachelor’s degree, I worked mostly on machine learning for particle physics and spent a couple of summers at CERN in Professor Maria Spiropulu’s group and Dr. Maurizio Pierini’s group.

Contact

vitoria.barin-pacela at mila.quebec (she/elle/ela)

Yoga

The practice of yoga inspires my life at every moment. I am finishing my yoga teacher training with Sylvie Tremblay at Yoga Sangha. I’m open to inquiries about classes, and I also would be interested in teaching yoga for AI communities, conferences and workshops.

On a research note, I’m interested in studying consciousness and working on the computational and mathematical fundamentals of yogic philosophy, with the purpose of better understanding and validating the tradition.

News

2026

Jul: I’m attending ICML in Seoul! I will present at the Workshop on Compositional Learning: Safety, Interpretability, and Agents. Looking forward to setting up meetings during the conference.
Aug: Our paper “Stop Probing, Start Coding: Why Linear Probes and Sparse Autoencoders Fail at Compositional Generalisation” was accepted at UAI. See you in Amsterdam!
Oct: Attending the BIRS workshop on Identifiable Representation Learning!

2025

Oct: Gave a workshop on VAEs at Bayes Plurinacional and gave a talk at EIA University in Colombia.
July: I’ll be attending ICML and I’ll present a poster at the SIM workshop!
June: I started an internship in the NeuroAI program at the Cold Spring Harbor Laboratory, working with Professor David Klindt.
May: I was a 2025 Upper Bound Talent Bursary recipient.
Winter: I am TAing for the Representation Learning course by Professor Aaron Courville.
Mar: Attending KHIPU in Santiago.
Feb: Attending the Bellairs Workshop on Causality in Barbados!

2024

Apr: Presenting a poster at CLeaR, LA, about our paper “On the Identifiability of Quantized Factors”!
Mar: I was awarded Mila’s EDI Excellence Scholarship.
Feb: Attended RIIAA in Quito.