QuantumLibraries icon indicating copy to clipboard operation
QuantumLibraries copied to clipboard

Published 20 hours ago • verified publisher icon microsoft

Hybrid QAOA

Open dlasecki opened this issue 4 years ago • 8 comments

This PR adds a library with the Quantum Approximate Optimization Algorithm (QAOA) written in Q#. It supports problems that can be encoded into 1-local and/or 2-local Hamiltonians. Additionally, it offers a hybrid QAOA algorithm in Q# and C# that uses a gradient-free Cobyla optimizer to choose optimal parameters for the QAOA in the form of a feedback loop. The hybrid QAOA supports interoperability with Python by a magic command. Values of parameters during the execution of the hybrid QAOA can be logged into a file.

dlasecki avatar Aug 08 '20 11:08 dlasecki

CLA assistant check
All CLA requirements met.

ghost avatar Aug 08 '20 11:08 ghost

Hi @dlasecki, thank you for your contribution! I was wondering if there is a design spec on this library that we could read? It would include but is not limited to introduction, problem statement, requirements, abstractions/design, out of scope/future features. This will greatly help our team review your PR and help maintain the library. LMK if you'd like to discuss, thanks!

guenp avatar Aug 10 '20 22:08 guenp

Hi @dlasecki, thank you for your contribution! I was wondering if there is a design spec on this library that we could read? It would include but is not limited to introduction, problem statement, requirements, abstractions/design, out of scope/future features. This will greatly help our team review your PR and help maintain the library. LMK if you'd like to discuss, thanks!

The relevant file shared through an email.

dlasecki avatar Aug 11 '20 13:08 dlasecki

Thanks @dlasecki, who did you email?

guenp avatar Aug 12 '20 00:08 guenp

Thanks @dlasecki, who did you email?

Yourself but not directly. :)

dlasecki avatar Aug 12 '20 10:08 dlasecki

@guenp Was the document that I shared with you helpful? In any case, I would be happy to join you all in a call to discuss the contribution, potential improvements or questions if you have any.

Regards, Dariusz.

dlasecki avatar Sep 17 '20 17:09 dlasecki

Thanks, @dlasecki for this contribution! Linking the original issue: https://github.com/microsoft/QuantumLibraries/issues/209

We are currently following up internally how to move forward with this PR. Thanks for your patience.

guenp avatar Nov 23 '20 17:11 guenp

Hi @guenp @cgranade,

I wanted to follow up on this PR and your internal discussions. Shall we discuss the big picture of how such optimization algorithms might be provided in Q# and possibly adjust/improve/extend this contribution to fit this vision or shall we think about delivering it to examples?

dlasecki avatar Jul 25 '21 07:07 dlasecki