conftrace_

2018 COLT COLT 2018

Open problem: Improper learning of mixtures of Gaussians

Abstract

We ask whether there exists an efficient unsupervised learning algorithm for mixture of Gaussians in the over-complete case (number of mixtures is larger than the dimension). The notion of learning is taken to be worst-case compression-based, to allow for improper learning.

Authors

Elad Hazan, Livni Roi

Topics

Machine Learning > Core Methods > Clustering Machine Learning > Learning Types > Unsupervised Learning Machine Learning > Optimization & Theory > Statistical Learning

Keywords

unsupervised learning mixture of gaussian improper learning over-complete case compression-based learning

More from COLT 2018

All COLT 2018 papers →

Geometric Lower Bounds for Distributed Parameter Estimation under Communication Constraints

Local moment matching: A unified methodology for symmetric functional estimation and distribution estimation under Wasserstein distance

Iterate Averaging as Regularization for Stochastic Gradient Descent

Smoothed analysis for low-rank solutions to semidefinite programs in quadratic penalty form

Certified Computation from Unreliable Datasets

Open Problem: The Dependence of Sample Complexity Lower Bounds on Planning Horizon