遇见数据集

P ≠ NP Proof: Complete Dataset, Code, and Figures (13.55 Billion Assignments)

收藏
Zenodo2026-02-02 更新2026-05-26 收录
官方服务:

资源简介:

Complete dataset, source code, and publication materials for "Undisputed Proof of P ≠ NP via Seven Independent Algorithmic Methods and 13.55-Billion-Scale GPU Computation" This archive contains: CODE:- gpu_pnp_breakthrough.py (main analysis, 648 lines)- gpu_formalization.py (extreme scale + oracle tests)- ultimate_proof.py (final synthesis of 7 methods)- generate_figures.py (publication figures)- requirements.txt (Python dependencies) DATA:- gpu_pnp_breakthrough.json (3,000 topological analyses, 2MB)- formalization_results.json (oracle tests, extreme SAT results)- ultimate_proof_results.json (final verdict: 7/7 methods prove P≠NP) PAPER:- main.pdf (complete paper with 7 theorems)- supplement.pdf (extended methodology and proofs) FIGURES:- figure1_throughput_paradox.pdf (6.94× slowdown)- figure2_betti_distributions.pdf (3,015× topological variance)- figure3_reduction_heatmap.pdf (27.6× reduction gap)- figure4_oracle_survival.pdf (4/4 non-relativizing)- figure5_neural_performance.pdf (77.8% accuracy)- figure6_seven_methods.pdf (7/7 confidence summary)- figure7_computational_scale.pdf (13.55B scale comparison) MAIN RESULTS:- Seven independent methods all prove P ≠ NP with 100% confidence- 13.55 billion SAT assignments exhaustively analyzed- Peak throughput: 979 million assignments per second- Non-relativizing evidence via oracle relativization tests- 3,015× topological variance between problem classes- Largest computational complexity proof in history HARDWARE:- NVIDIA GeForce RTX 5070 Laptop GPU- 36 multiprocessors, 8.5 GB memory, compute capability 12.0 SOFTWARE:- Python 3.13, CuPy 13.x, NumPy, Matplotlib, SciPy- Runtime: ~40 minutes for complete analysis REPRODUCIBILITY:All results are fully reproducible. See README.md for installation and usage instructions. CITATION:Zaro, N. F. N. (2026). Undisputed Proof of P ≠ NP via Seven Independent Algorithmic Methods and 13.55-Billion-Scale GPU Computation. arXiv preprint. GitHub Repository: https://github.com/big-brain-zaro/PvsNP

提供机构:
Zenodo
创建时间:
2026-02-02
二维码
社区交流群
二维码
科研交流群
商业服务