Optimization.TrustRegion.Nesterov1983

Nesterov's Optimal Gradient method

optimalGradient

Arguments

:: (Additive f, Functor f, Ord a, Floating a, Epsilon a)

=> a

condition number, kappa

-> a

Lipschitz constant, l

-> (f a -> f a)

gradient of function

-> a

initial step size, alpha0

-> f a

starting point

-> [f a]

iterates

Nesterov 1983 optimalGradient kappa l df alpha0 x0 is Nesterov's optimal gradient method, first described in 1983. This method requires knowledge of the Lipschitz constant l of the gradient, the condition number kappa, as well as an initial step size alpha0 in (0,1).