confidence_snn.m

神经网络的工具箱, 神经网络的工具箱,

function [ylc, yuc, y_av] = confidence_snn(c_point_conf, nets, alpha, P)
%CONFIDENCE_SNN Estimate confidence intervals
%
%  Syntax
%
%    [ylc, yuc, y_av] = confidence_snn(c_conf, nets, alpha, P)
%
%  Description
%
%    CONFIDENCE_SNN takes
%      c_conf    - c_confidence.
%      nets      - [1 x M] net_structs of trained networks with cost
%                  function WCF_SNN.
%      alpha     - [1 x M] matrix of network weighting factors.
%      P         - [N0 x MU] matrix with inputs
%    and returns
%      ylc       - lower bound for confidence interval
%      yuc       - upper bound for confidence interval
%      y_av      - weigthed average output
%
%   (N0 = #inputs; MU = #patterns; M = #networks in ensemble)
%
%  Algorithm
%
%    See: T. Heskes; Practical confidence and prediction Intervals.
%    Advances in Neural Information Processing Systems 9, pages
%    176-182, Cambridge, 1997, MIT Press.
%
%  See also
%
%    C_CONFIDENCE_SNN, BALANCE_SNN
%

errf = nets(1).costFcn.fn;

[y_av, conf_err_estimate] = simff_avr_snn(nets, alpha, P);
[NL, MU] = size(y_av);
e_max = conf_err_estimate .* repmat(c_point_conf, 1, MU);

[ylc, yuc] = inverse_error_snn(y_av, e_max, errf);