Is sqrt d Separation Necessary for Gradient EM to Learn Gaussian Mixtures in High Dimensions?

stat.ML, cs.LG

Submitted: 2026-10-06

Updated: 2026-10-06

Comments: 51 pages

Code: https://github.com/Ryanzhang887/EM_Separation

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Implicit High-Order Moment Tensor Estimation and Learning Latent Variable Models
• Global Convergence of Gradient EM for Over-Parameterized Gaussian Mixtures

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• Efficient quadratic entropy with distance sketches
• Embedding-Bias in Conditional Independence Testing
• From Log-Odds to Shapley Values: An Explanatory Geometry for the Weighted Naive Bayes Classifier