Sciweavers

Share
ICANN
2001
Springer

Fast Curvature Matrix-Vector Products

10 years 4 months ago
Fast Curvature Matrix-Vector Products
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 a number of ways to adopt ideas from conjugate gradient in the stochastic setting, using fast Hessian-vector products to obtain curvature information cheaply. In our benchmark experiments the resulting highly scalable algorithms converge about an order of magnitude faster than ordinary stochastic gradient descent.
Nicol N. Schraudolph
Added 29 Jul 2010
Updated 29 Jul 2010
Type Conference
Year 2001
Where ICANN
Authors Nicol N. Schraudolph
Comments (0)
books