Function-space VI
Approximate inference directly over functions, keeping the emphasis on uncertainty and predictive structure.
Academic Profile
Luis A. Ortega
Postdoctoral Researcher · Aalborg University
Probabilistic Machine Learning for uncertainty, inference, and generalization.
Function-space variational inference, linearized Laplace approximations, deep ensembles, and Chernoff-style generalization bounds.
Bayesian deep learning
Variational inference
Generalization bounds
Deep ensembles
10
Publications
5
Ongoing projects
4
Research lines
Current Position
Postdoctoral research at Aalborg University.
I am a Postdoctoral Researcher in the Department of Computer Science at Aalborg University, working in the Section for Distributed, Embedded and Intelligent Systems. My work connects probabilistic machine learning and uncertainty quantification with the section's broader focus on distributed, embedded, and intelligent systems.
Research Scope
Linearized Laplace
Post-hoc Bayesian uncertainty for modern neural networks without forcing every page into technical detail.
PAC-Chernoff bounds
A path into the generalization-theory work for expert readers and reviewers.
Deep ensembles
Calibration, diversity, and model combination presented as connected research rather than isolated papers.
Selected Work
ESANN 2026 · Spotlight talk
Scalable Linearized Laplace Approximation via Surrogate Neural Kernel
Learns a surrogate neural kernel to avoid large Jacobians while estimating uncertainty for pre-trained networks.ICML 2024
Variational Linearized Laplace Approximation for Bayesian Deep Learning
Approximates LLA via sparse variational Gaussian processes with sub-linear training costs.JAIR · ECAI 2025 Spotlight