Rydberg simulations

Optimizations of pulse sequences for quantum gates applied on neutral atom arrays using the Rydberg blockade mechanism.

Installation:

Clone or download the repository and install via pip:

$ pip install -e ./ --user

Usage:

Begin by importing the module.

from rydberg import System

Create a rydberg-system-object, choosing what gate to optimize over (implemented ones are CNOT, CZ, Toffoli/CCNOT and CCZ). Default parameters are chosen if no parameters are given.

params = {
    "operation": "CNOT",
    "pulse_sequence": "CNOT",
    "opt_algorithm": "BOBYQA",
    "density": false,
    "num_of_tests": 0,
    "num_of_evaluations": 20000,

    "print_every": 1,
    "verbose": false,
    "save_data": false
    }

rbs = System("CNOT", parameters=params)

Next, load and set up the inputs and targets of the optimization, and then optimize.

rbs.build_codes()
rbs.optimize()

Both rows can also be run as one as rbs.build_codes().optimize(). When the optimization is done, the results can be printed and plotted with the display_results function.

rbs.display_results()

The plot will show the pulses applied from left to right, where the heights of the bars correspond to the strength of each respective pulse (in $\pi$), and the numbers above the bars corresponds to the qubit number (in a row).

Default parameters are stored in parameters.json, the different gates are stored in codes.json and the pulse sequence templates are stored in pulse_sequences.json.