Victor Gallego

I’m a data scientist and predoctoral researcher at the ICMAT (Spanish National Research Council) in Madrid and I study foundational topics and applications in Machine Learning & Data Science. In more detail, I focus on mixing Bayesian statistics with ML models in order to achieve more robust and secure decision-making systems. I'm doing my PhD under the supervision of David Ríos Insua (ICMAT, Royal Academy of Sciences) and David Gómez-Ullate (ICMAT, UCA). Previously, I did a double degree in Mathematics and Computer Science and a Masters in Mathematical Engineering at UCM. Welcome to my website! I expect to populate it soon!

News

I'll be a Visiting Research Scholar at Duke University (Statistics Department) and SAMSI (The Statistical and Applied Mathematical Sciences Institute) from August to December 2019!

Publications (selected and peer-reviewed, full list @ Google Scholar)

Variationally Inferred Sampling Through a Refined Bound.
Victor Gallego and David Rios Insua.
Advances in Approximate Bayesian Inference (AABI 2019). paper and code.
Stochastic Gradient MCMC with Repulsive Forces.
Victor Gallego and David Rios Insua.
NIPS 2018 Worksop on Bayesian Deep Learning. arxiv and poster.
Reinforcement Learning under Treats.
Victor Gallego, Roi Naveiro and David Rios Insua.
AAAI 2019. Student Track. arxiv and code.
Assessing the effect of advertising expenditures upon sales: A Bayesian structural time series model.
Victor Gallego, Pablo Suárez García, Pablo Angulo and David Gómez-Ullate.
Applied Stochastic Models in Business and Industry (Q1). 2019. arxiv and poster.
Bayesian Factorization Machines for Risk Management and Robust Decision Making.
Pablo Angulo, Victor Gallego, David Gómez-Ullate and Pablo Suárez García.
Mathematical and Statistical Methods for Actuarial Sciences and Finance. 2018. paper.

Teaching

Spring 2019: Machine Learning with R (15 h).
Targeted to researchers from the Instituto Nacional de Estadística (National Statistics Institute). Jointly with Alberto Torres, Roi Naveiro and David Rios Insua. course materials
Fall 2018: Stochastic Processes (30 h).
Problems/practical sessions. Targeted to last year undergraduate students in Maths & Stats and Math & Econ at Complutense University in Madrid. Jointly with Antonio Gómez Corral. solved class exams

Talks

Variationally Inferred Sampling for Probabilistic Programs.
Bayesian Inference in Stochastic Processes (BISP 2019). slides and poster.
Markov Decision Processes under Threats.
Advances in Decision Analysis (ADA 2019). slides and poster.

CV

A short CV can be downloaded here.