Approximates a gradient by finite differences.
A line search optimizes a function of one variable without analytic gradient information.
Implements the Backtracking Linesearch like that in LBFGS-C (which is (c) 2007-2010 Naoaki Okazaki under BSD)
A diff function that supports subsets of the data.
Represents a differentiable function whose output is guaranteed to be consistent
The empirical hessian evaluates the derivative for multiplcation.
The Fisher matrix approximates the Hessian by E[grad grad'].
Port of LBFGS to Scala.
A line search optimizes a function of one variable without analytic gradient information.
Anything that can minimize a function
Implements the Orthant-wise Limited Memory QuasiNewton method, which is a variant of LBFGS that handles L1 regularization.
Represents a function for which we can easily compute the Hessian.
SPG is a Spectral Projected Gradient minimizer; it minimizes a differentiable function subject to the optimum being in some set, given by the projection operator projection
A differentiable function whose output is not guaranteed to be the same across consecutive invocations.
Minimizes a function using stochastic gradient descent
Implements a TruncatedNewton Trust region method (like Tron).
Implements the L2^2 and L1 updates from Duchi et al 2010 Adaptive Subgradient Methods for Online Learning and Stochastic Optimization.
Class that compares the computed gradient with an empirical gradient based on finite differences.