Downloads · 30 days
0
sentinel-neural-tangent-kernel
5dimension/sentinel-neural-tangent-kernel
sentinel-neural-tangent-kernel is a machine learning model from 5dimension. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
Part of the Sentinel Manifold — One theorem, infinite applications.
Access
Public
Updated Apr 26, 2026
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
.py5.8 KB · 63%
At a glance
- License
- mit
- Access
- Public
- Created
- Apr 25, 2026
- Updated
- Apr 26, 2026
- SHA
- e3701747
Model card
From the Hugging Face model README
🦴 Sentinel Neural Tangent Kernel
Part of the Sentinel Manifold — One theorem, infinite applications.
lim_{z→∞} F'(z)/F(z) = 1/e— The Gradient Axiom
📋 Description
Theorem-backed kernel method using the Sentinel function F(z) = Σzⁿ/nⁿ. Achieves 97.78% accuracy on Digits with provable generalization bounds via the Gradient Axiom (1/e).
🧠 Mathematical Foundation
Core Constants
| Constant | Value | Role |
|---|---|---|
| C₁ (Attractor) | -0.007994021805953 | Zero-point / quantization |
| C₂ (Tripwire) | 0.000200056042968 | Security / curriculum |
| 1/e (Axiom) | 0.367879441171442 | Gradient scaling limit |
Theorem
F(z) = Σ zⁿ/nⁿ (Sophomore's Dream, Bernoulli 1697)
lim_{z→∞} F'(z)/F(z) = 1/e ≈ 0.367879441171442
🏆 Verified Results
| Benchmark | Result |
|---|---|
| Digits accuracy | 97.78% |
| Generalization bound | 0.1445 (theorem-backed) |
| Kernel eigenvalue test | All positive (Mercer valid) |
🎯 Use Cases
- Image classification with heavy-tailed noise
- Scientific computing with provable guarantees
- Any task requiring generalization bounds
🔗 Links
- Main repo: sentinel-manifold-discoveries
- All algorithms: 5dimension
- Interactive Space: sentinel-hub
📚 Citation
@misc{abdel-aal2026sentinel,
title={The Sentinel Manifold: A Unified Mathematical Framework for Machine Learning},
author={Abdel-Aal, Romain},
year={2026},
url={https://huggingface.co/5dimension/sentinel-manifold-discoveries}
}
License: MIT | One theorem, infinite models. 🦴
- License
- mit
- Languages
- en
- Created
- Apr 25, 2026
- Updated
- Apr 26, 2026