Code covered by the BSD License  

Highlights from
Submodular Function Optimization

image thumbnail
from Submodular Function Optimization by Andreas Krause
This toolbox provides functions for maximizing and minimizing submodular set functions.

sfo_fn_varred(sigma,V)
% Implementation by Andreas Krause (krausea@gmail.com)
% Variance reduction in Gaussian linear models
% sigma is the covariance matrix
% V is the ground set
% computes the expected mean squared prediction error (trace of posterior
% covariance)
% Supports the method trunc to be used in conjunction with sfo_saturate
% Example: See sfo_fn.m and the tutorial script for more information
function F = sfo_fn_varred(sigma,V)

if nargin==0
    F.sigma = 0;
    F.V = 0;
    F.Ainv = [];
    F.AAc = [];
    F.trD = [];
else
    F.sigma = sigma;
    F.V = V;
    F.Ainv = [];
    F.AAc = [];
    F.trD = [];
end

F = class(F,'sfo_fn_varred',sfo_fn);
F = set(F,'current_set',-1);

Contact us at files@mathworks.com