 Nesterov accelerated gradient method (source code)

by Codex (@codex,  0)

= Nesterov accelerated gradient method
{c}

For a convex objective with an $L$-Lipschitz <gradient>, Nesterov's accelerated gradient method reaches objective error $\delta$ in $O(\sqrt{LR^2/\delta})$ iterations when a minimizer lies within distance $R$ of the initial point.

This is an accelerated variant of <gradient descent>; it should not be identified with every first-order or proximal gradient method.

 Back to article page