The fixed angle QAOA

The fixed angle conjecture for the quantum approximate optimization algorithm onregular MaxCut graphs

The universal angles

The universal angles for MaxCut on regular graphs QAOA with guaranteed performance are located in angles_regular_graphs.json

File structure

connectivity (str):
    p (str):
        gamma: list
        beta: list
        AR: float # The approximation ratio guaranteed by the angles

Usage

def get_angles(p: int, conn: int) -> tuple:
    try:
        data_p = data[str(conn)][str(p)]
        gamma, beta = data_p['gamma'], data_p['beta']
        return gamma, beta
    except KeyError:
        print('Angles not found!')

Directory structure

The general approach is to have the data separate from the presentation, à-la MVC pattern.

• data/*.(nc|json|npy) - data files
• data/generators/ - scripts that generate the data
• plots/*.ipynb - scripts that generate figures
• plots/pdf/ - pdf output from figure generators

Nice tools used

You'll need these to understand and run the code

• QTensor - tensor network simulator with focus on MaxCut
• Cartesian Explorer - a handy tool to map multi-dimensional data
• xarray - used to store datasets with their coordinates in netcdf format
• Gurobi and gurobipy - classical optimization of combinatorial problems.

Gurobi installation

See instructions on the official site, the general steps are the following:

1. Register at the website gurobi.com
2. Unpack an archive
3. Run licensing code (included in data/generators/*)