This program is designed to observe the effects of meta-learning on QAOA problems. The QAOA solver code is an expansion of a quantum Variational Monte Carlo program written by Tianchen Zhao, Giuseppe Carleo, James Stokes, and Shravan Veerapaneni, which may be seen here: https://github.com/Ericolony/QAOA
Meta-learning is done by an implementation of the Reptile algorithm developed by Nichol, Achiam, and Schulman at OpenAI.
So far this code is able to solve MaxCut and Ising Model QAOA problems using the Python package NetKet.
This repository is no longer being regularly updated; while the code contained here should be functional, there are no guarantees that any one part of it will work as intended.