Question on "newlind" from Neural Network Toolbox

3 views (last 30 days)
Dear community,
I would like to know what algorithm exactly (LMS ?) describes the matlab estimation function newlind. http://www.mathworks.de/de/help/nnet/ref/newlind.html
The documentation of Matlab does not offer full disclosure on the function, but only what it does. More info very appreciated.
Kind regards

Accepted Answer

Greg Heath
Greg Heath on 22 Sep 2014
Use type, doc or help to get the answer. For example,
type newlind
function out1 = newlind(varargin)
%NEWLIND Design a linear layer.
%
% newlind(X,T) takes an RxQ input matrix X, an SxQ target matrix T, and
% returns a linearlayer designed to output T (with minimum sum square
% error) given X.
%
% newlind(X,T,Xi) can also solve for linear networks with input delays and
% multiple inputs and layers by supplying input and target time series
% data in cell array form:
% X - NixTS cell array, each element X{i,ts} is an RixQ input matrix.
% T - NtxTS cell array, each element T{i,ts} is an VixQ matrix.
% Xi - NixID cell array, each element Xi{i,k} is an RixQ matrix, default = [].
% returns a linear network with ID input delays, Ni network inputs, Nl layers,
% and designed to output T (with minimum sum square error) given input P.
Now try the doc and help commands.
Hope this helps.
Thank you for formally accepting my answer
Greg

More Answers (3)

Greg Heath
Greg Heath on 14 Sep 2014
Use the type command:
type newlind
Hope this helps
Thank you for formally accepting my answer
Greg

Erol H
Erol H on 21 Sep 2014
Thanks, but i would like to know if the newlind command uses the LMS (Least mean squares) algorithm or not ?

Star Strider
Star Strider on 21 Sep 2014
In the newlind documentation under More About > Algorithms:
  • newlind calculates weight W and bias B values for a linear layer from inputs P and targets T by solving this linear equation in the least squares sense:
[W b] * [P; ones] = T
So the answer to your question is: Yes. ‘Least squares’ means ‘least sum of squares’ that is essentially least mean squares. (It doesn’t take the mean but the result is essentially the same.)

Categories

Find more on Sequence and Numeric Feature Data Workflows in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!