How do i find the eigenvalues of this tough Hamiltonian?

Click For Summary
SUMMARY

The discussion focuses on finding the eigenvalues of a complex Hamiltonian represented as a 6D tensor, where each G matrix resides in a different space. The user is exploring methods to convert this tensor into a matrix format suitable for eigenvalue computation. Suggestions include constructing a matrix with specific index arrangements and considering tensor contraction techniques. The challenge lies in the compatibility of the G matrices due to their differing spaces, which complicates direct multiplication.

PREREQUISITES
  • Understanding of Hamiltonian mechanics and quantum systems
  • Familiarity with eigenvalue problems in linear algebra
  • Knowledge of tensor algebra and contraction methods
  • Experience with matrix representation of multi-dimensional data
NEXT STEPS
  • Research methods for tensor contraction in quantum mechanics
  • Learn about constructing matrices from multi-dimensional tensors
  • Study eigenvalue algorithms applicable to non-standard matrix forms
  • Explore software tools for symbolic computation, such as Mathematica or MATLAB
USEFUL FOR

Quantum physicists, applied mathematicians, and researchers working on complex Hamiltonians or tensor analysis in quantum systems.

baouba
Messages
40
Reaction score
0
I have this Hamiltonian --> (http://imgur.com/a/lpxCz)

Where each G is a matrix.

I want to find the eigenvalues but I'm getting hung up on the fact that there are 6 indices. Each G matrix lives in a different space so I can't just multiply the G matrices together. If I built this Hamiltonain explicitly on the computer it would be a 6D tensor right? So how do I change it into a matrix so I can find the eigenvalues?

One thing I've been told it that I could build a matrix with the i' indices going down and the i indices going across. Neglecting the sum over the alphas, if I did this would the first row of my matrix be (G1,1 G1,1G1,1,G1,1 G1,1 G1,2,G1,1 G1,1 G1,3,...,G1,1 G1,1 G1,10,G1,1 G1,2 G1,1,...,G1,1 G1,2 G1,10,...,G1,1 G1,10 G1,10,G1,2 G1,1G1,1,...G1,10 G1,10 G1,10)
where each element is the product of the elements of the G matrices. I don't see you can take the product of the elements of the G matrices if they aren't defined in the same space. Would constructing a matrix like this really give the eigenvalues of H?
 
Physics news on Phys.org
The other thing I'm thinking might work is to use some sort of tensor contraction. Is there a way to do this? I'm not sure how to approach this problem, any help would be greatly appreciated!
 

Similar threads

  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 1 ·
Replies
1
Views
719
  • · Replies 0 ·
Replies
0
Views
3K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 18 ·
Replies
18
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 6 ·
Replies
6
Views
944
  • · Replies 7 ·
Replies
7
Views
873