[英]How to add Tikhonov regularization in “scipy.optimize.least_squares”?
As the title, could anyone tell me a way to add "Tikhonov regularization" in "scipy.optimize.least_squares"?作为标题,谁能告诉我在“scipy.optimize.least_squares”中添加“Tikhonov 正则化”的方法? The default seems L2.
默认似乎是 L2。
A straightforward way is to add more rows to your design matrix and objective vector.一种直接的方法是向您的设计矩阵和目标向量添加更多行。
In other words if you have A and b, and seek x to minimise换句话说,如果你有 A 和 b,并寻求 x 以最小化
||A*x-b||
and you change A and b to你把 A 和 b 改成
A~ = (A)
(K)
b~ = (b)
(k)
then the least squares problem: minimise那么最小二乘问题:最小化
||A~*x - b~||
is minimising正在最小化
||A*x-b||^2 + ||K*x-k||^2
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.