Small Covers for Near-zero Sets of Polynomials and Learning Latent Variable Models
Offered By: IEEE via YouTube
Course Description
Overview
Explore a 25-minute IEEE conference talk on advanced statistical techniques for learning latent variable models. Delve into Gaussian Mixture Models, parameter estimation, and the Method of Moments. Examine the challenges of low degree moments and dimension reduction. Investigate the use of higher moments and zero sets in overcoming obstacles. Understand the main technical lemma and its application to cover and density estimation. Learn about mixtures of linear regressions and their implications. Gain insights from experts Ilias Diakonikolas (UW Madison) and Daniel Kane (UCSD) on small covers for near-zero sets of polynomials and their relevance to learning latent variable models.
Syllabus
Intro
Gaussian Mixture Models
Parameter Estimation
Method of Moments
Low Degree Moments
Obstacle
Dimension Reduction Redux
Using Higher Moments
Zero Set
Difficulties
Main Technical Lemma
Cover
Density Estimation
Mixtures of Linear Regressions
Conclusions
Taught by
IEEE FOCS: Foundations of Computer Science
Tags
Related Courses
Introduction to Artificial IntelligenceStanford University via Udacity Natural Language Processing
Columbia University via Coursera Probabilistic Graphical Models 1: Representation
Stanford University via Coursera Computer Vision: The Fundamentals
University of California, Berkeley via Coursera Learning from Data (Introductory Machine Learning course)
California Institute of Technology via Independent