遇见数据集

Data and code for: The fixed-N first moment of common link-completions for a spacelike anchor pair in a causal diamond

收藏
Zenodo2026-09-24 更新2026-10-01 收录
官方服务:

资源简介:

Code, frozen protocols, raw results and verification scripts supporting the paper "The fixed-N first moment of common link-completions for a spacelike anchor pair in a causal diamond" (Paper I of the fixed-anchor moment programme). The deposit contains:- the frozen overlap-kernel and importance-sampling code, and the direct-resprinkling comparator, byte-identical to the versions bound by both prospective confirmation freezes;- the predeclared protocols and the frozen benchmark report (kernel fixtures, gates, injected-error test, volume benchmark);- both prospective confirmations: the first nine-cell panel, which did not meet its equivalence criterion for lack of precision, and the held-out six-cell panel reported in the paper — with their freezes, per-cell results and analyses;- the development phase that separated them, including the common-offset experiment and the thirty-run standard-error calibration;- the sampled-population kernel audit and the general-anchor checks;- the anchor-representativeness measurement with both of its primary sources, as a self-verifying sub-archive;- four verification scripts that recompute the paper's results from this deposit, and the review records made while the paper was written, including corrections to earlier claims. Every file is listed with its SHA-256 hash in MANIFEST.json. Scripts in the verification/ folder run from the unpacked deposit; the other scripts are archived exactly as they ran, for audit. Requirements: Python 3.10+, NumPy, SciPy, mpmath. From the unpacked paper_I_data folder: python verification/factcheck_paper_I.py recomputes every number in the paper's results section python verification/compare_confirmations.py shows both confirmations bound identical code python verification/verify_kernel_derivation.py rebuilds the overlap kernel from the paper's text python verification/cst_context_numbers.py discreteness scale; binomial vs Poisson size The anchor-representativeness sub-archive (representativeness/overshoot_deposit.tar.gz) carries its own README and verification script.

提供机构:
Zenodo
创建时间:
2026-09-24
二维码
社区交流群
二维码
科研交流群
商业服务