Stochastic Cubic Regularization for Fast Nonconvex Optimization

This paper proposes a stochastic variant of a classic algorithm—the cubic-regularized Newton method [Nesterov and Polyak, 2006]. The proposed algorithm efficiently escapes saddle points and finds approximate local minima for general smooth, nonconvex functions in only Õ( −3.5) stochastic gradient and stochastic Hessian-vector product evaluations. The latter can be computed as efficiently as stochastic gradients. This improves upon the Õ( −4) rate of stochastic gradient descent. Our rate matches the bestknown result for finding local minima without requiring any delicate acceleration or variance-reduction techniques.