4 papers
Angle-free cluster robust Ritz value bounds for restarted block eigensolvers
Ming Zhou, Andrew V. Knyazev, Klaus Neymeyr
Convergence rates of block iterations for solving eigenvalue problems typically measure errors of Ritz values approximating eigenvalues. The errors of the Ritz values are commonly…
Convergence analysis of a block preconditioned steepest descent eigensolver with implicit deflation
Ming Zhou, Zhaojun Bai, Yunfeng Cai +1
Gradient-type iterative methods for solving Hermitian eigenvalue problems can be accelerated by using preconditioning and deflation techniques. A preconditioned steepest descent it…
Majorization-type cluster robust bounds for block filters and eigensolvers
M. Zhou, M. E. Argentati, A. V. Knyazev +1
Convergence analysis of block iterative solvers for Hermitian eigenvalue problems and the closely related research on properties of matrix-based signal filters are challenging, and…
Exponential convergence rates for Batch Normalization: The power of length-direction decoupling in non-convex optimization
Jonas Kohler, Hadi Daneshmand, Aurelien Lucchi +3
Normalization techniques such as Batch Normalization have been applied successfully for training deep neural networks. Yet, despite its apparent empirical benefits, the reasons beh…