Reproducibility materials for "Quantum Oracle Search versus Evolutionary Metaheuristics on Molecular Landscapes: An Access-Count Benchmark"
Year: 2026
Type: software
Abstract: Manuscript source and compiled PDFs, Supplementary Information, simulation and analysis programs, cached molecular landscapes, JSON run artifacts and the analysis ledger for a benchmark of the Durr-Hoyer oracle-search construction against local search, genetic algorithms, simulated annealing, covariance-matrix adaptation, differential evolution and particle swarm optimisation on 22 entries covering 21 distinct rigid torsional landscapes (N up to 65536), plus synthetic control families up to 2^24, with one unit charged per objective access. The manuscript reports no statistically distinguishable oracle advantage on the real molecular entries (access ratios 0.86 to 6.03; the one sub-unity point, serine at 0.86, has a selection-aware 95% interval spanning parity). v1.0.3 ships the current manuscript tree (ASOC260924/) with its gate suite, and adds the runs behind the Supplementary Information's selection-aware intervals for all 22 entries, the stereochemistry and torsion-subset sensitivity reruns, and the size-sweep panel. README.md lists the commands to rerun each result and states which figures and tables are regenerated by a program and which are not.
Author: TCHAPET NJAFA, Jean-Pierre
Institution: Université de Yaoundé I
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7212016602