performing a least squares with regularisation in matlab

Question

cgo 2018-7-13

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/410165-performing-a-least-squares-with-regularisation-in-matlab

评论： Sterling Baird 2020-9-23

I have data sets X (2n by 8) and Y(2n by 1). I want to find the coefficients a so that Y = Xa. So we can perform a = X\Y (as a least squares minimisation).

I wanted to ask if it possible to proceed with a form of regularisation (L1 or something simple) from this?

Please help.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Diwakar 2018-7-13

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/410165-performing-a-least-squares-with-regularisation-in-matlab#answer_328698

My understanding of your problem is that you want to find the coefficient a. So in order to implement optimization you can implement average of sum of least squares as shown below.

Loss= ((Y-X*a)'*(Y-X*a))/(2*n);

The above shown function is a vectorized implementation of the squared error loss function. So this can be minimized in order to get the optimal value of a. If you want to fit a curve to this then any form of regularization should be fine.

Hope this helps

Cheers!

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Sterling Baird 2020-9-23

How would you actually implement the regularization though?

请先登录，再进行评论。

Answer 2

Bruno Luong 2020-9-23

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/410165-performing-a-least-squares-with-regularisation-in-matlab#answer_499270

编辑：Bruno Luong 2020-9-23

Simpless method:

n = size(X,2); % 8
lambda = 1e-6; % <= regularization parameter, 0 no regularization, larger value stronger regularized solution 
a = [X; lambda*eye(n)] \ [Y; zeros(n,1)]