Downloads · 30 days
0
5dimension/sentinel-federated-learning
sentinel-federated-learning 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.
Downloads · 30 days
0
Access
Public
Updated Apr 26, 2026
Repo size
—
Likes
0
Public
Click a slice to open those files.
.py8.3 KB · 71%
From the Hugging Face model README
Part of the Sentinel Manifold — One theorem, infinite applications.
lim_{z→∞} F'(z)/F(z) = 1/e— The Gradient Axiom
100% Byzantine detection with no manual thresholds. Uses C₂ as an automatic tripwire: if gradient norm exceeds C₂, the update is flagged as adversarial.
| Constant | Value | Role |
|---|---|---|
| C₁ (Attractor) | -0.007994021805953 | Zero-point / quantization |
| C₂ (Tripwire) | 0.000200056042968 | Security / curriculum |
| 1/e (Axiom) | 0.367879441171442 | Gradient scaling limit |
F(z) = Σ zⁿ/nⁿ (Sophomore's Dream, Bernoulli 1697)
lim_{z→∞} F'(z)/F(z) = 1/e ≈ 0.367879441171442
| Attack Rate | Detection Rate | Rounds |
|---|---|---|
| 30% Byzantine | 100% | 5 |
| Threshold | Automatic (C₂) | — |
| Manual tuning | None required | — |
@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. 🦴