[edit]
Learning mixtures of Gaussians with maximum-a-posteriori oracle
Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, PMLR 15:489-497, 2011.
Abstract
We consider the problem of estimating the parameters of a mixture of distributions, where each component distribution is from a given parametric family e.g. exponential, Gaussian etc. We define a learning model in which the learner has access to a “maximum-a-posteriori” oracle which given any sample from a mixture of distributions, tells the learner which component distribution was the most likely to have generated it. We describe a learning algorithm in this setting which accurately estimates the parameters of a mixture of $k$ spherical Gaussians in $\mathbb{R}^d$ assuming the component Gaussians satisfy a mild separation condition. Our algorithm uses only polynomially many (in $d, k$) samples and oracle calls, and our separation condition is much weaker than those required by unsupervised learning algorithms like [Arora 01, Vempala 02].