1: \begin{abstract}

2:     %This paper proposes the first Stochastic Tensor Method for convex optimization. This method is based on the minimization of the regularized Taylor polynomial with sub-sampled derivatives.  In the case of the third-order scheme our method requires $O(\e^{-2})$ computations of stochastic gradients, $O(\e^{-4/3})$  computations of stochastic Hessians, and none computations of stochastic third derivatives. The convergence rate matches the classical (non-accelerated) tensor method.

3:     We propose a general non-accelerated tensor method under inexact information on higher-order derivatives, analyze its convergence rate, and provide sufficient conditions for this method to have similar complexity as the exact tensor method. As a corollary, we propose the first stochastic tensor method for convex optimization and obtain sufficient mini-batch sizes for each derivative.  

4:     %\pd{Motivated by stochastic convex optimization methods exploiting higher-order derivatives, we propose a general non-accelerated tensor method under inexact information on higher-order derivatives, analyze its convergence rate, and provide sufficient conditions for this method to have similar complexity as the exact tensor method. As a corollary, we propose a stochastic tensor method and obtain sufficient mini-batch sizes for each derivative such that the resulting method with high probability has the same complexity as the exact tensor method.}   

5: \end{abstract}

6: