A collection of data and software used in J. Tuziemski, J. Pawłowski, P. Tarasiuk, Ł. Pawela and B. Gardas, Recent quantum runtime (dis)advantages.
Ising model-like instances with 3-body interactions that serve as structured benchmarks for higher-order optimization problems. We use:
- single SWAP and Cauchy distributed couplings (S2q = 1, S3q = 6),
- single SWAP and symmetrized Pareto distributed couplings.
For each type and size N ∈ {80, 100, 130, 156}, 50 random instances are provided. Interactions are described in CSV format. We also include Python code used for their generation, and reference energies used for optimality gap calculations presented in the paper.
Python client that collects experimental data on methods based on SBM (Simulated Bifurcation Machine), using VeloxQ SDK. Combined results are subject to simple statistical processing, in order to compute measures such as success fractions and TTε (time needed to reach a specific optimality gap).
Julia project that contains our minimal implementation of Simulated Annealing capable of handling instances with 3-body interactions, such as HUBO_instances. As described in the paper, the algorithm evaluates the flip costs for each processed variable. The proposed implementation relies on CUDA.jl and uses optimized CUBLAS and CUSPARSE calls for selected linear algebra operations.
Python program used to configure and run a quantum circuit implemented on the native gate set of IBM Brisbane.
The first stage is preprocessing, which involves translating a quantum circuit into the topology and native gate set of a chosen device, a process known as transpilation. The transpiled circuit is then executed multiple times (shots), with consecutive runs separated by readout and thermalization.
Julia workspace for solving Simon's problem on selected oracles using classical computing. It implements multiple pseudo-random Oracle functions (periodic or not), a universal solver, and a solver for special scenarios when the number of bits set in a period is restricted. Each solver, in addition to solving the Simon problem, returns the period when one is found.
Collection of all the data points presented on the plots included in the paper in CSV files. Python code used to generate the plots is included as well, alongside the plot-specific options described in JSON files that are loaded by the scripts.