MHB How to Translate Univariate Newton's Approximation to Multidimensional Systems?

Click For Summary
SUMMARY

The discussion focuses on translating univariate Newton's approximation into a multidimensional context, specifically for the equation system y = F.x, where y and x are (nx1) column vectors and F is an (nxn) coefficient matrix. The initial attempt to express the multidimensional approximation as x2 = x1 - (F.x1 / F'.x1) was corrected by clarifying that F' represents a matrix of derivatives, leading to the formulation x^{(k+1)} = x^{(k)} - (DF(x^{(k)})^{-1}·F(x^{(k)}), where DF(x) is the matrix of derivatives. This approach allows for solving systems of equations involving multiple variables.

PREREQUISITES
  • Understanding of Newton's method for root-finding
  • Familiarity with matrix calculus and derivatives
  • Knowledge of linear algebra concepts, particularly matrix inversion
  • Basic understanding of vector notation and operations
NEXT STEPS
  • Study the derivation and application of the Jacobian matrix in multidimensional calculus
  • Learn about the implications of matrix inversion in numerical methods
  • Explore the convergence criteria for Newton's method in higher dimensions
  • Investigate alternative methods for solving nonlinear systems, such as the Broyden's method
USEFUL FOR

Mathematicians, data scientists, and engineers working with optimization problems, particularly those involving multidimensional systems and numerical methods for solving nonlinear equations.

Purplepixie
Messages
7
Reaction score
0
Hello,
I am having difficulty in translating the univariate Newton's approximation {Xn = Xn-1 - [ f(Xn-1) / f'(Xn-1)]} into the multidimensional case. My multidimensional equation system is y = F.x where y and x are (nx1) column vectors and the coefficients matrix F is (nxn), so that (nx1) = (nxn).(nx1) = (nx1).

I have translated the univariate Newton's approximation to the n-variate case as:
x2 = x1 - (F.x1 / F'.x1) <=> x1 - (F.x1).Inv(F'.x1)

But F'.x1 is a nx1 column vector and has no inverse. I then thought that perhaps the x1's cancel out . But if so then we would have x2 = x1 - F.Inv(F') with the last term a nxn matrix, so (nx1) = (nx1) - (nxn), which is not possible.

Your assistance would be greatly appreciated!

(This is my first post incidentally, so pls excuse any breaches of protocol)
 
Physics news on Phys.org
Purplepixie said:
Hello,
I am having difficulty in translating the univariate Newton's approximation {Xn = Xn-1 - [ f(Xn-1) / f'(Xn-1)]} into the multidimensional case. My multidimensional equation system is y = F.x where y and x are (nx1) column vectors and the coefficients matrix F is (nxn), so that (nx1) = (nxn).(nx1) = (nx1).

I have translated the univariate Newton's approximation to the n-variate case as:
x2 = x1 - (F.x1 / F'.x1) <=> x1 - (F.x1).Inv(F'.x1)

But F'.x1 is a nx1 column vector and has no inverse. I then thought that perhaps the x1's cancel out . But if so then we would have x2 = x1 - F.Inv(F') with the last term a nxn matrix, so (nx1) = (nx1) - (nxn), which is not possible.

Your assistance would be greatly appreciated!

(This is my first post incidentally, so pls excuse any breaches of protocol)

Hi Purplepixie, welcome to MHB!

First we have to establish what $F'$ is.
If it is a matrix that does not depend on any variable, then $F'$ is the matrix that contains only zeroes.
Consequently we won't really get any useful result.

Newton's approach aims to solve $f(x)=0$, and makes use of the fact that $f(x) \approx f(x_0) + f'(x)(x-x_0) \implies x_0\approx x - \frac{1}{f'(x)}\cdot f(x)$.
We might try to solve $F(x_1,\ldots,x_n)=(0,\ldots, 0)$ in a similar fashion. In this case $F$ is not a matrix, but a set of $n$ functions. And each function has $n$ parameters.
If we take the derivative, we don't have just 1 derivative, but instead we have the derivatives of $n$ functions with respect to each of the $n$ parameters.
The result is a matrix of $n\times n$ derivatives. Let's call it $DF(x)$ to denote that it's a matrix of functions that depend on $x$.
Now we can write Newton's approximation as $x^{(k+1)} = x^{(k)} - \Big(DF(x^{(k)})\Big)^{-1}\cdot F(x^{(k)})$, where $x^{(k)}$ denote the successive approximations of $x$.
 
Dear Klaas,
Thank you for your clear and succinct explanation. I can now see where I was wrong. So much of mathematics would be simplified if representational systems could be improved!
All the best,
PP
 
I am studying the mathematical formalism behind non-commutative geometry approach to quantum gravity. I was reading about Hopf algebras and their Drinfeld twist with a specific example of the Moyal-Weyl twist defined as F=exp(-iλ/2θ^(μν)∂_μ⊗∂_ν) where λ is a constant parametar and θ antisymmetric constant tensor. {∂_μ} is the basis of the tangent vector space over the underlying spacetime Now, from my understanding the enveloping algebra which appears in the definition of the Hopf algebra...

Similar threads

  • · Replies 1 ·
Replies
1
Views
4K
Replies
2
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K