activity
20082026
most citedExponential convergence rates for Batch Normalization: The power of length-direction decoupling in non-convex optimization

28 citations · 72 across the 9 of their papers we have counts for

collaborators
Showing 2011Show all

1 paper · 1 filter

math.NA2011★ 2 cited

A geometric convergence theory for the preconditioned steepest descent iteration

Klaus Neymeyr

Preconditioned gradient iterations for very large eigenvalue problems are efficient solvers with growing popularity. However, only for the simplest preconditioned eigensolver, name…