Open anedumla opened 4 years ago
Hi, this sounds interesting. I hope we could have a discussion on this topic.
Hi, this sounds interesting. I hope we could have a discussion on this topic.
Good to hear that! yes, of course, we can talk about it.
My name is Atsuya Hasegawa. I am interested in it.
I’m in
I'm interested in this! Computer science major
I'm in
I am interested in this issue.
Abstract
The problem of Hamiltonian simulation can be described as, given a hermitian matrix H and a real number t, find a circuit for exp(iHt).
Usually this is a hard problem and a widely used algorithm is:
Step 2. is already implemented in Qiskit. The goal of this project would be to obtain the full algorithm by implementing Step 1., which would consist on two parts: finding a decomposition and finding circuits for each exp(iH_j t).
One of the existing ideas to find a decomposition is to view H as the adjacency matrix of a graph, find an edge colouring of such graph and associate the subgraph induced by colour j to H_j. Such a decomposition guarantees that the H_j are 1-sparse (at most one non-zero entry per row or column), and in the literature it has been described how to obtain the circuits for the Hamiltonian simulation of 1-sparse matrices.
Refereneces:
Description
Members
@slackhandle
email:example@example.com
Deliverable
GitHub repo