mlpack  git-master
GreedyDescent Class Reference

Greedy descent policy for Stochastic Co-ordinate Descent(SCD). More...

Static Public Member Functions

template
<
typename
ResolvableFunctionType
>
static size_t DescentFeature (const size_t, const arma::mat &iterate, const ResolvableFunctionType &function)
 The DescentFeature method is used to get the descent coordinate for the current iteration. More...

 

Detailed Description

Greedy descent policy for Stochastic Co-ordinate Descent(SCD).

This descent scheme picks a the co-ordinate for the descent with the maximum guaranteed descent, according to the Gauss-Southwell rule. This is a deterministic approach and is generally more expensive to calculate.

For more information, refer to the following.

@misc{Nutini2015,
author = {Julie Nutini and Mark Schmidt and Issam H.
Laradji and Michael Friedlander and Hoyt Koepke},
title = {Coordinate Descent Converges Faster with the Gauss-Southwell Rule
Than Random Selection},
year = {2015},
eprint = {arXiv:1506.00552}
}

Definition at line 38 of file greedy_descent.hpp.

Member Function Documentation

◆ DescentFeature()

static size_t DescentFeature ( const size_t  ,
const arma::mat &  iterate,
const ResolvableFunctionType &  function 
)
inlinestatic

The DescentFeature method is used to get the descent coordinate for the current iteration.

Template Parameters
ResolvableFunctionTypeThe type of the function to be optimized.
Parameters
iterationThe iteration number for which the feature is to be obtained.
iterateThe current value of the decision variable.
functionThe function to be optimized.
Returns
The index of the coordinate to be descended.

Definition at line 53 of file greedy_descent.hpp.


The documentation for this class was generated from the following file:
  • /var/www/www.mlpack.org/mlpack-git/src/mlpack/core/optimizers/scd/descent_policies/greedy_descent.hpp