Jean-Baptiste Tristan





I am a machine learning researcher. I work as a consulting member of technical staff at Oracle Labs in Boston. I am a senior member of the ACM. A long time ago I worked on computer science topics that were not machine learning.

Research Interests

Safe Machine Learning

My main research interest is safe machine learning. This covers common topics such as fairness, accountability, specifications, and assurance.

Bayesian Machine Learning

I'm interested in the design of Bayesian inference algorithms. One of the key challenge in Bayesian inference algorithms is that there is often a complex trade-off between computational and statistical properties. For example, marginalizing a random variable in a model might improve convergence of a Gibbs sampler, but at the cost of making the sampler sequential. In my work I consider statitistical performance in light of computational complexity, parallelism, and memory footprint.

Scalable Machine Learning

Some of my work focuses on scaling machine learning algorithms to large datasetes and large models either by using randomized methods for dimentionality reduction or parallelization and distribution.

Talks Highlights


Service Highlights