Research

My research program focuses on probabilistic AI: how to build learning and decision systems that remain accurate, interpretable, and reliable under uncertainty.

Thematic areas #

1) Amortized inference and GFlowNets #

I study the theory and practice of GFlowNets and related generative models, with emphasis on training objectives, stability under balance violations, and correctness assessment.

2) Bayesian modeling and prior predictive methods #

I develop principled methods for prior specification and identifiability, especially for matrix and tensor factorization models. A central question is how prior predictive behavior should guide model design.

3) Sequential decision-making, causality, and trustworthy AI #

I work on uncertainty-aware approaches for sequential learning and causal discovery, including human-in-the-loop workflows and reliability-oriented evaluation.

4) Application-driven probabilistic ML #

I collaborate on problems in recommender systems, language technologies, and scientific modeling where calibrated uncertainty improves decisions and deployment safety.

Collaborators #

I actively collaborate with colleagues at the University of Coimbra, BCAM, Fundação Getulio Vargas (EMAp), and international research groups.

For publication details, see the Publications section.