Minimizing error function -> sloving linear system of equations

Click For Summary
SUMMARY

The discussion focuses on minimizing an error function to estimate coefficients \(\alpha_{uv}\) for a function \(f(x,y)\) through a linear system of equations. The error function is defined as \(E(\alpha)=\sum_{x,y}w(x,y)\Big( f(x,y)-\sum_{u,v=-N}^{N}\alpha_{u,v}f(x+u, y+v)\Big)^{2}\), where \(w(x,y)\) is a weighted probability function. The solution involves setting the derivative \(\frac{\partial E}{\partial \alpha_{s,t}}=0\) to derive a linear system, with Gaussian elimination identified as the most efficient general method for solving such systems. The discussion also highlights the need for MATLAB functions to facilitate this process.

PREREQUISITES
  • Understanding of error functions and their minimization techniques.
  • Familiarity with linear algebra concepts, particularly linear systems of equations.
  • Proficiency in MATLAB for implementing numerical solutions.
  • Knowledge of Gaussian elimination and its application in solving equations.
NEXT STEPS
  • Research MATLAB functions for solving linear systems, such as 'linsolve' or 'mldivide'.
  • Study Gaussian elimination with pivoting for efficient equation solving.
  • Explore advanced optimization techniques for error minimization in numerical analysis.
  • Learn about the implementation of weighted functions in error calculations.
USEFUL FOR

Students and professionals in mathematics, engineering, and data science who are involved in numerical analysis, optimization problems, and MATLAB programming for solving linear systems.

Theraven1982
Messages
23
Reaction score
0

Homework Statement


I want to find certain coefficients [tex]\alpha_{uv}[/tex] by minimizing a error function. In the end, I want to make a function of this, so iteration is not a problem. I want to estimate these coefficients to find the best estimate for [tex]f(x,y)[/tex].



Homework Equations


The error function:
[tex] E(\alpha)=\sum_{x,y}w(x,y)\Big( f(x,y)-\sum_{u,v=-N}^{N}\alpha_{u,v}f(x+u, y+v)\Big)^{2}[/tex]
with
[tex] w(x,y)=P(f(x,y)\in M_1|f(x,y)),[/tex]
which in turn is equal to

[tex] \frac{1}{\sigma\sqrt{2\pi}} \frac{exp\Big[ -\frac{1}{2\sigma^2} \Big( f(x,y) - \sum_{u,v=-N}^{N}\alpha_{u,v}f(x+u, y+v)\Big)^2\Big]}{ \sum_{i=1}^{2}P(f(x,y)|f(x,y)\in M_{i}) }[/tex]

The Attempt at a Solution


[tex] \frac{\partial E}{\partial \alpha_{s,t}}=0[/tex]

eventually gives:

[tex] \sum_{u,v=-N}^{N}\alpha_{u,v}=\frac{ \sum_{x,y}w(x,y)f(x+s, y+t)f(x,y) }{ \sum_{x,y}w(x,y)f(x+s, y+t)f(x+u, y+v) }[/tex]

So, now we have a linear system of equations. I don't know how to solve this in a efficient way.
Thanks in advance,
 
Last edited:
Physics news on Phys.org
That's a very wide question! There are many different ways of solving a linear system- which is most efficient depends on the coefficients. Gaussian eliminaion with pivoting is probably the best general method.
 
I was afraid of that ;). I'll find my linear algebra book ;).
Is there anyone who has experience in solving these equations in Matlab? There are probably functions for this purpose ; again, any kick in the right direction is welcome ;)
 

Similar threads

Replies
4
Views
1K
Replies
5
Views
2K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
3
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
Replies
8
Views
2K
  • · Replies 27 ·
Replies
27
Views
2K
Replies
2
Views
1K