1: \begin{abstract}

2: 	We propose a parallel stochastic Newton method (PSN) for minimizing unconstrained smooth convex functions. We analyze the method in the strongly convex case, and give conditions under which acceleration can be expected when compared to its serial counterpart. We show how PSN can be applied to the empirical risk minimization problem, and demonstrate the practical efficiency of the method through numerical experiments and models of simple matrix classes. 

3: 	

4: %		We present a novel parallel minimization algorithm that is inspired by Stochastic Dual Coordinate Ascent (SDNA) which we call Parallel Random Block Newton Method (PRBNM). Our method uses curvature information as its serial counterpart to improve the convergence rate over the pure first-order methods. It has superior convergence properties than the serial version, and allows for efficient implementation. The convergence of PRBNM is superior to Parallel Coordinate Descend Method in both theory and practice as is demonstrated in our analysis.

5: 		

6: 		% \PACS{PACS code1 \and PACS code2 \and more}

7: 		% \subclass{MSC code1 \and MSC code2 \and more}

8: 	\end{abstract}

9: