Stochastic Trust Region Inexact Newton Method for Large-scale Machine Learning
This addresses optimization efficiency for large-scale machine learning practitioners, though it appears incremental as it builds on existing stochastic second-order methods.
The authors tackled the problem of large-scale machine learning optimization by proposing STRON, a stochastic trust region inexact Newton method that uses progressive subsampling and conjugate gradient solvers. Their empirical results demonstrate the method's efficacy against existing approaches on benchmark datasets.
Nowadays stochastic approximation methods are one of the major research direction to deal with the large-scale machine learning problems. From stochastic first order methods, now the focus is shifting to stochastic second order methods due to their faster convergence and availability of computing resources. In this paper, we have proposed a novel Stochastic Trust RegiOn Inexact Newton method, called as STRON, to solve large-scale learning problems which uses conjugate gradient (CG) to inexactly solve trust region subproblem. The method uses progressive subsampling in the calculation of gradient and Hessian values to take the advantage of both, stochastic and full-batch regimes. We have extended STRON using existing variance reduction techniques to deal with the noisy gradients and using preconditioned conjugate gradient (PCG) as subproblem solver, and empirically proved that they do not work as expected, for the large-scale learning problems. Finally, our empirical results prove efficacy of the proposed method against existing methods with bench marked datasets.