
Introducing PROMISE, a suite of sketching-based preconditioned stochastic gradient algorithms for fast convergence on ill-conditioned large-scale convex optimization problems in machine learning.


coremsg

preconditioned-stochastic-optimization-methods-for-ill-conditioned-large-scale-convex-optimization-problems-in-machine-learning


Preconditioned Stochastic Optimization Methods for Ill-Conditioned Large-Scale Convex Optimization Problems in Machine Learning


title_rewrite


提案されたアルゴリズムは、機械学習における大規模な凸最適化問題において高速な収束を実現する。


promise-preconditioned-stochastic-optimization-methods-by-incorporating-scalable-curvature-estimates


PROMISE: Preconditioned Stochastic Optimization Methods by Incorporating Scalable Curvature Estimates
