The Program for Economic Research (PER)’s Fall Mini Course will be held in two parts.
Part I: Thursday, September 26, 2:30pm– 4:30pm, Room 301, Uris Hall (get directions)
Part II: Friday, September 27, 2:00pm-4:00pm, Room 301, Uris Hall (get directions).
*Open to current Columbia University students only
David Blei, Professor of Statistics & Computer Science, Columbia University
A core problem of modern statistics and machine learning is to approximate difficult-to-compute probability distributions. This problem is especially important in probabilistic modeling and Bayesian statistics, which frame all inference about unknown quantities as calculations about conditional distributions.
This tutorial reviews variational inference (VI), a method that approximates probability distributions through optimization. VI has been used in myriad applications in machine learning and tends to be faster than more traditional methods, such as Markov chain Monte Carlo sampling.
After discussing the basics of variational inference, Professor Blei will describe some of the pivotal tools for VI that have been recently developed: Monte Carlo gradients, black box variational inference, stochastic variational inference, and variational autoencoders. The Mini Course will conclude with a review of some of the unsolved problems in VI and promising research directions.
The Mini Course is open to all current Columbia University students. Register here.