Gonçalo Faria

bayesian deep learning, deep generative modelling, nlp, causal discovery


Hey, I'm Gonçalo.

After spending some time in industry, I am currently an AI researcher at Instituto de Telecomunicações, working with André Martins on LLMs. I am preparing to embark on a PhD journey in Computer Science at the University of Washington, under the guidance of Professor Noah Smith.

My research interests are robustness to distribution shifts, o.o.d generalization, and reliability, particularly for complex reasoning tasks.

I studied at the University of Minho for my undergrad and at Instituto Superior Técnico for my master's. My master's dissertation, advised by André Martins and Mário Figueiredo, focused on Differentiable Causal Discovery.

Feel free to reach out! I'm always happy to chat about research, life, or anything else.

CLEAR 2022
Gonçalo R. A. Faria, André F. T. Martins, Mário A. T. Figueiredo
Abstract : Recent work has shown promising results in causal discovery by leveraging interventional data with gradient-based methods, even when the intervened variables are unknown. However, previous work assumes that the correspondence between samples and interventions is known, which is often unrealistic. We envision a scenario with an extensive dataset sampled from multiple intervention distributions and one observation distribution, but where we do not know which distribution originated each sample and how the intervention affected the system, \textit{i.e.}, interventions are entirely latent. We propose a method based on neural networks and variational inference that addresses this scenario by framing it as learning a shared causal graph among an infinite mixture (under a Dirichlet process prior) of intervention structural causal models. Experiments with synthetic and real data show that our approach and its semi-supervised variant are able to discover causal relations in this challenging scenario.

Also on Google Scholar
Honors & awards
APRP Prize - Best Master Thesis 2022
APRP establishes the APRP prize for the best master's thesis, with the purpose of distinguishing work of high merit in the area of pattern recognition in Portugal.
Professor Luís Vidigal Prize - Best Master Thesis 2022
It is intended to annually award Masters of any master's degree at IST, authors of the best Master's Thesis whose theme falls within the scientific areas of Electrical and Computer Engineering and Informatics.
Almedina Prize 2020
Awarded annually to the best scholarship student of each Organic Teaching and Research Unit.
Scholarship Calouste Gulbenkian Foundation 2018
New Talents in Artificial Intelligence, 2nd Edition.