Stochastic Variance-Reduced Cubic Regularized Newton Method

02/13/2018
by   Dongruo Zhou, et al.
0

We propose a stochastic variance-reduced cubic regularized Newton method for non-convex optimization. At the core of our algorithm is a novel semi-stochastic gradient along with a semi-stochastic Hessian, which are specifically designed for cubic regularization method. We show that our algorithm is guaranteed to converge to an (ϵ,√(ϵ))-approximately local minimum within Õ(n^4/5/ϵ^3/2) second-order oracle calls, which outperforms the state-of-the-art cubic regularization algorithms including subsampled cubic regularization. Our work also sheds light on the application of variance reduction technique to high-order non-convex optimization methods. Thorough experiments on various non-convex optimization problems support our theory.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/31/2019

Stochastic Recursive Variance-Reduced Cubic Regularization Methods

Stochastic Variance-Reduced Cubic regularization (SVRC) algorithms have ...
research
07/07/2020

A Distributed Cubic-Regularized Newton Method for Smooth Convex Optimization over Networks

We propose a distributed, cubic-regularized Newton method for large-scal...
research
10/12/2022

A Momentum Accelerated Adaptive Cubic Regularization Method for Nonconvex Optimization

The cubic regularization method (CR) and its adaptive version (ARC) are ...
research
12/11/2018

On the Ineffectiveness of Variance Reduced Optimization for Deep Learning

The application of stochastic variance reduction to optimization has sho...
research
09/27/2022

Approximate Secular Equations for the Cubic Regularization Subproblem

The cubic regularization method (CR) is a popular algorithm for unconstr...
research
09/26/2018

Stochastic Second-order Methods for Non-convex Optimization with Inexact Hessian and Gradient

Trust region and cubic regularization methods have demonstrated good per...
research
08/21/2023

A Homogenization Approach for Gradient-Dominated Stochastic Optimization

Gradient dominance property is a condition weaker than strong convexity,...

Please sign up or login with your details

Forgot password? Click here to reset