David M. Blei

Prof. Blei and his group develop novel models and methods for exploring, understanding, and making predictions from the massive data sets that pervade many fields.  Their work is widely used in science, scholarship, and industry to solve interdisciplinary, real-world problems. In particular, they focus on a variety of applications, including language, recommendation systems, neuroscience, and the computational social sciences.  Prof. Blei and his group have set new paths in the fields of machine learning and artificial intelligence.

By bringing together ideas in computer science, statistics, and optimization, more than a decade ago, Blei and collaborators developed a method to discover the abstract “topics” that pervade a collection of documents. Today, their algorithm—latent Dirichlet allocation (LDA)—is a standard method for topic discovery, and is used in many downstream tasks.  Since then, Blei and his group has significantly expanded the scope of topic modeling. One recent example is collaborative topic models, which connect textual content to user behavior (such as clicks), and which can be used to interpret patterns of readership, recommend documents, characterize readers, and organize collections according to both content and consumption.  (This algorithm is used by the New York Times to form recommendations for its readers.)  In addition to working on topic models, Blei and his group have created generic algorithms for scaling a wide class of statistical models to massive data sets.  Their work on variational inference has changed the scale at which we can apply sophisticated methods for data science and machine learning.

Blei earned his Bachelor's degree in Computer Science and Mathematics from Brown University (1997) and his PhD in Computer Science from the University of California, Berkeley (2004). Before joining Columbia, he was an Associate Professor of Computer Science at Princeton University (2006-2014).

Research Areas

• Natural Language Processing and Speech
• Computational Biology
• Applied and Theoretical Machine Learning
• Artificial Intelligence
• Causal Inference
• Generative AI & Large Language Models (LLMs)