#machine learning
Jun 2025
Global Convergence of Gradient EM for Over-Parameterized Gaussian Mixtures
This is the first global convergence and recovery result for EM or Gradient EM beyond the special case of m=2, and it is proved that with only mild over-parameterization, randomly initialized gradient EM converges to the ground truth with polynomial time and samples.
Mo Zhou, Weihang Xu, Maryam Fazel et al.
· arXiv.org · 2 citations
· ⚡1