Max Balandat

Research Scientist

I lead the Modeling & Optimization team within the Adaptive Experimentation group on Facebook’s Core Data Science team. We focus on developing methods and tools for probabilistic modeling and sample-efficient optimization, and apply them to a broad range of applications across the company, including infrastructure optimization, AutoML, online A/B tests, ranking systems, and AR/VR. I also lead the development of BoTorch, an open-source library for Bayesian Optimization in PyTorch.

In the past, I have also worked on the intersection of Machine Learning and Econometrics, in particular on how to utilize Machine Learning algorithms to perform causal inference in experimental and non-experimental settings.

I hold an MA in Mathematics and a PhD in Electrical Engineering and Computer Sciences from UC Berkeley.


Bayesian Optimization, Gaussian processes, probabilistic modeling, causal inference and machine Learning

Latest Publications

NeurIPS - December 7, 2020

BOTORCH: A Framework for Efficient Monte-Carlo Bayesian Optimization

Max Balandat, Brian Karrer, Daniel Jiang, Sam Daulton, Ben Letham, Andrew Gordon Wilson, Eytan Bakshy

NeurIPS - December 7, 2020

Efficient Nonmyopic Bayesian Optimization via One-Shot Multi-Step Trees

Shali Jiang, Daniel Jiang, Max Balandat, Brian Karrer, Jacob R. Gardner, Roman Garnett