Stabilizer codes

Implementation of stabilizer codes in pyQuil

CS269Q Spring 2018-19 course project

Authors: Shubham Chandak, Jay Mardia, Meltem Tolunay

Instructor: Will Zeng

Report

Installation

Follow instructions to install qvm and quilc from http://docs.rigetti.com/en/stable/. Then just run

git clone https://github.com/shubhamchandak94/stabilizer_code

to download the code.

Defining a stabilizer code

The stabilizer codes are defined here in terms of the stabilizer generator matrix and the error correction capabilities of the code. The error correction capability is defined in terms of a dictionary mapping the error type to the number of such errors the code can correct. Currently supported error types are X, Y, Z on single qubits and a combination of X and Z on different qubits. Some examples of stabilizer codes along with the parameters are provided in stabilizer_check_matrices.py.

Generating programs for encoding and decoding a stabilizer code

Given a stabilizer code representation my_stabilizer_code in the format as described above, run the following

import stabilizer_code
code = stabilizer_code.StabilizerCode(my_stabilizer_code)

Then code.encoding_program and code.decoding_program are the encoding and decoding programs, respectively.

Running basic tests

First run qvm -S and quilc -S before running any tests.

Several standard stabilizer matrices are defined in stabilizer_check_matrices.py

To run basic tests (bit flip, phase flip etc. with different initial states) for all codes in above file, run python3 basic_tests.py

To run basic tests for a specific code, run with the name of the code, e.g., python3 basic_tests.py bit_flip_code

Plotting logical error rates vs physical error rates

First run qvm -S and quilc -S before running any tests. Then run the function logical_vs_physical_error_rates.py after changing the parameters in the main appropriately (examples provided in the code).

Plotting logical error rates vs code rates

First run qvm -S and quilc -S before running any tests. Then run the function PlotCodeRateVsErrRate from code_rate_vs_logical_error_rate.py with appropriate arguments (examples provided in the code)