Combining Conjugate Direction Methods with Stochastic Approximation of Gradients Export

@misc{citeulike:2736643, abstract = {The method of conjugate directions provides a very eective way to optimize large, deterministic systems by gradient descent. In its standard form, however, it is not amenable to stochastic approximation of the gradient.}, author = {Schraudolph, N. and Graepel, T.}, citeulike-article-id = {2736643}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.12.9715}, keywords = {conjugategradient, minibatch, onlinelearning, sgd, stochasticgradientdescent}, posted-at = {2008-04-30 08:58:53}, priority = {3}, title = {Combining Conjugate Direction Methods with Stochastic Approximation of Gradients}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.12.9715}, year = {2002} }

TY - GEN ID - citeulike:2736643 L3 - citeulike-article-id:2736643 N2 - The method of conjugate directions provides a very eective way to optimize large, deterministic systems by gradient descent. In its standard form, however, it is not amenable to stochastic approximation of the gradient. TI - Combining Conjugate Direction Methods with Stochastic Approximation of Gradients KW - conjugategradient KW - minibatch KW - onlinelearning KW - sgd KW - stochasticgradientdescent AU - Schraudolph, N AU - Graepel, T PY - 2002/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.12.9715 ER -