Paper (Zenodo report): https://zenodo.org/records/18816815

XPRIZE Quantum Applications Benchmark Harness

This repository is the technical support repo for an XPRIZE Quantum Applications Phase I Wild Card submission focused on rare-event conjunction-risk estimation (Pc) under uncertainty. It contains reproducible classical baselines (MC / IS / SPLIT), realistic stress regimes (mixtures + near-threshold boundary layers), an oracle-oriented encounter-plane model, and Phase II evidence artifacts (precision calibration + telescoping/gating + QDK resource-estimator outputs).

Quickstart

python -m venv .venv
source .venv/bin/activate && python -m pip install -U pip && pip install -r requirements.txt
python scripts/run_encounterplane_replicates.py && python scripts/plot_encounterplane_paper_figures.py && python scripts/run_splitting_suite.py && python scripts/plot_ciwidth_vs_compute.py && python scripts/generate_resource_estimate.py && python scripts/run_qdk_resource_estimate.py --bits 16,24,32 --error_budgets 1e-2,1e-3,1e-4 && python scripts/run_gating_simulation.py && python scripts/plot_gating_results.py && python scripts/run_precision_sweep.py && python scripts/plot_precision_sweep.py

Windows activation equivalent: .venv\\Scripts\\activate

Repo Map

• src/xq/: core models and estimators (encounter model, encounter-plane model, distributions/scenarios, IS, splitting, quantization).
• scripts/: runnable experiment/plot/summary entrypoints.
• results/: generated outputs (tables, plots, markdown summaries); generally treated as generated artifacts.
• paper/: tracked, judge-facing figure pack used directly in narrative/PDF assembly.

Reproduce Key Results

Encounter-plane suite + paper figures

python scripts/run_encounterplane_replicates.py && python scripts/plot_encounterplane_paper_figures.py

Rare-event baselines (SPLIT + compute-normalized comparison)

python scripts/run_splitting_suite.py && python scripts/plot_ciwidth_vs_compute.py

Viability resources (oracle decomposition + scaling tables/plots)

python scripts/generate_resource_estimate.py

QDK resource estimator sweep

python scripts/run_qdk_resource_estimate.py --bits 16,24,32 --error_budgets 1e-2,1e-3,1e-4

Phase II plan evidence (telescoping + precision calibration)

python scripts/run_gating_simulation.py && python scripts/plot_gating_results.py
python scripts/run_precision_sweep.py && python scripts/plot_precision_sweep.py

How This Maps to the Narrative

• Impact: gating/telescoping evidence quantifies that only a minority of cases need expensive refinement.
• Benchmarking: MC, IS, SPLIT are compared on single, mixture, and near-threshold regimes.
• Viability: oracle sketch, scaling table, and QDK estimates provide transparent resource assumptions.
• Phase II plan: calibrated precision targets and gray-zone routing define a concrete hybrid insertion path for QAE.

Script Index

Benchmarks

Script	Purpose	Key outputs
scripts/run_bench_suite.py	Tiered MC/IS benchmark summary	results/tables/bench_suite_summary.csv
scripts/plot_ciwidth_across_tiers.py	MC/IS CI-width across tiers	results/plots/ciwidth_across_tiers.png
scripts/run_splitting_suite.py	SPLIT replicate benchmark	results/tables/splitting_suite.csv
scripts/plot_ciwidth_vs_compute.py	MC/IS/SPLIT CI vs compute	results/plots/ciwidth_vs_compute.png

Encounter-plane

Script	Purpose	Key outputs
scripts/run_encounterplane_replicates.py	Replicate CIs for MC/IS on encounter-plane scenarios	results/tables/encounterplane_replicates.csv
scripts/plot_encounterplane_paper_figures.py	Judge/paper figure set (baseline + near-threshold)	results/plots/encounterplane_paper_baseline.png, results/plots/encounterplane_paper_nearthreshold.png
scripts/generate_encounterplane_summary.py	Encounter-plane narrative summary	results/encounterplane_summary.md

Viability + QDK

Script	Purpose	Key outputs
scripts/generate_resource_estimate.py	Parametric FT viability tables/plots + oracle note	results/tables/resource_scaling_table.csv, results/plots/ae_vs_mc_calls.png, results/oracle_spec.md
scripts/run_qdk_resource_estimate.py	QDK-based arithmetic skeleton resource sweep	results/tables/qdk_resource_estimate_summary.csv, results/plots/qdk_resource_estimate.png, results/qdk_resource_estimate.md

Phase II evidence

Script	Purpose	Key outputs
scripts/run_gating_simulation.py	Telescoping policy simulation over scenario population	results/tables/gating_simulation.csv
scripts/plot_gating_results.py	Gating policy figures + summary markdown	results/plots/gating_*.png, results/gating_summary.md, results/phase2_plan_evidence.md
scripts/run_precision_sweep.py	Quantization precision sweep with replicate MC	results/tables/precision_sweep.csv
scripts/plot_precision_sweep.py	Precision error/CI plots + bit recommendation summary	results/plots/precision_*_vs_bits.png, results/precision_sweep_summary.md

Cite This Work

• Paper DOI: 10.5281/zenodo.18816815
• Software/archive DOI: 10.5281/zenodo.18816743
• Technical report PDF: paper/technical_report_v0.1.pdf
• Release tag: v0.1.0

License

Source code is released under the MIT License. Research narrative, figures, reports, and generated result artifacts are included for competition/reproducibility context; cite the Zenodo records above when referencing the submission.