Joaquin Alvarez

Thank you for visiting my personal website! I am interested in rigorous statistical tools and frameworks for reliable machine learning, particularly in distribution-free uncertainty quantification and predictive inference. More broadly, I am excited about the idea of building trust in predictive algorithms in a black-box manner, with a formal data-driven perspective. I studied applied mathematics at ITAM in Mexico. I was fortunate to be guided by Miguel Angel Mota, Mauricio Romero and Edgar Francisco Roman-Rangel.

Blog posts

2024 ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization

Had the opportunity to participate in EAAMO Conference, 2024. This year’s edition took place in beautiful San Luis Potosi, Mexico. Loved the talks focusing on ML fairness. Everyone a...

Neuroscience workshop in Trieste

It was an amazing experience to attend to the neuroscience workshop! I’ve been fascinated to discover connections between how the human brain learns and how machines learn. I was im...

Predicting crimes in Boston

This is joint work with my colleagues Diego Velazquez and Marcelino Sanchez. This is a summary of a project work for the Advanced Regression Analysis course at ITAM. In this blog...

Reflections about Bentkus inequality

Photo by Seyfettin Dincturk. In honor to Vidmantas Bentkus. Context In 2005 Vidmantas Bentkus introduced a novel probability concentration inequality. Astonishingly, this inequa...

Predicting traffic in the New York City Subway

This is joint work with Julieta Rivero and Rodrigo Villela. In this blog post you can find a summary of our project at ITAM where we used linear regression models to make predict...