In this paper we study the second-order optimality of decentralized stochastic algorithm that escapes saddle point efficiently for nonconvex optimization problems.

As hyper-parameters are ubiquitous and can significantly affect the model performance, hyper-parameter optimization is extremely important in machine learning.

Work completed at the University of Washington. Work completed at Microsoft Research.

Riemannian setting is a highly challenging affair.

Zhang et al. have recently shown that geodesic convex concave Riemannian problems always admit saddle-point solutions.

Finally, we prove that the algorithm's

Finally, we prove that the algorithm's