Numerical recipes in C (2nd ed.): the art of scientific computing
Numerical recipes in C (2nd ed.): the art of scientific computing
Fast exact multiplication by the Hessian
Neural Computation
Dynamics and algorithms for stochastic search
Dynamics and algorithms for stochastic search
Fast curvature matrix-vector products for second-order gradient descent
Neural Computation
Stable Adaptive Momentum for Rapid Online Learning in Nonlinear Systems
ICANN '02 Proceedings of the International Conference on Artificial Neural Networks
Universal parameter optimisation in games based on SPSA
Machine Learning
The factored policy-gradient planner
Artificial Intelligence
Towards adjusting mobile devices to user's behaviour
MSM'10/MUSE'10 Proceedings of the 2010 international conference on Analysis of social media and ubiquitous data
| Hi-index | 0.00 |
The method of conjugate gradients provides a very effective way to optimize large, deterministic systems by gradient descent. In its standard form, however, it is not amenable to stochastic approximation of the gradient. Here we explore ideas from conjugate gradient in the stochastic (online) setting, using fast Hessian-gradient products to set up low-dimensional Krylov subspaces within individual mini-batches. In our benchmark experiments the resultingonline learningalg orithms converge orders of magnitude faster than ordinary stochastic gradient descent.