>[!warning]
>This content has not been peer reviewed.
# Standard Model — RST calibration results
**Summary:** The RST fidelity function was calibrated against running coupling data from the PDG for three Standard Model forces: QCD ($n = 8$), QED ($n = 1$), and the weak force ($n = 3$). The gauge-constrained fits produce R² = 0.96 (QCD), 0.69 (QED), and 0.16 (Weak). A discovery mode with free $n$ yields R² > 0.99 for QCD but finds fitted $n$ values far below the gauge generator counts, revealing the **RST-logarithmic boundary**: the algebraic fidelity function cannot replicate the logarithmic running of perturbative QFT.
---
## 1. Model
All fits use the additive RST coupling model:
$\alpha(Q) = \alpha_0 + A \cdot \left[\frac{1}{\mu(Q/\Lambda,\, n)} - 1\right]$
For QED, the observable is $1/\alpha$ (which decreases with $Q$, making it structurally equivalent to $\alpha_s$).
**Benchmark:** 1-loop perturbative QCD ($\alpha_s = 12\pi / [(33 - 2n_f)\ln(Q^2/\Lambda^2)]$) and 1-loop QED vacuum polarization (electron loops only).
**Data:** PDG 2024 compilation, 11 QCD points (1.78 GeV – 1 TeV), 5 QED points (Thomson limit – Z pole), 5 weak mixing angle points (0 – 91.2 GeV).
---
## 2. QCD (SU(3), n = 8)
### Constrained fit (n = 8)
| Parameter | Value |
|:---|---:|
| $\alpha_0$ | 0.1163 |
| A | 0.00015 |
| $\Lambda$ | 2545 GeV |
| **R²** | **0.959** |
| RMSE | 0.016 |
### 1-loop benchmark
| Parameter | Value |
|:---|---:|
| $\Lambda_{QCD}$ | 0.129 GeV |
| **R²** | **0.989** |
| RMSE | 0.008 |
### Interpretation
The RST model with $n = 8$ achieves R² = 0.96, a meaningful fit given that only 3 free parameters are used against logarithmic data. The optimizer pushes $\Lambda$ to 2545 GeV — far above the physical $\Lambda_{QCD} \approx 0.2$ GeV — because with $n = 8$, the fidelity transition is extremely sharp. The only way to spread the curve across 3 decades of $Q$ is to place $\Lambda$ above the data range, so all data falls in the gradual approach to the transition.
The 1-loop perturbative formula (R² = 0.989) wins because the actual running is logarithmic: $\alpha_s \propto 1/\ln(Q/\Lambda)$. The RST model produces algebraic (power-law) running: $\alpha_s \approx \alpha_0 + A \cdot (\Lambda/Q)^8/8$ at high $Q$. The $1/Q^8$ tail decays far too rapidly compared to $1/\ln(Q)$.
![[standard_model_qcd.png]]
*QCD running coupling: RST (n=8, red) vs 1-loop (blue dashed) vs PDG data (black dots).*
---
## 3. QED (U(1), n = 1)
### Constrained fit (n = 1, on 1/α)
| Parameter | Value |
|:---|---:|
| $C_0$ | 130.96 |
| A | 0.557 |
| $\Lambda$ | 0.006 GeV |
| **R²** | **0.694** |
| RMSE | 1.61 |
### 1-loop benchmark (electron loops only)
| Parameter | Value |
|:---|---:|
| **R²** | **−0.62** |
| RMSE | 3.71 |
### Interpretation
The RST model outperforms the 1-loop QED formula, but this comparison is misleading: the 1-loop formula only includes electron vacuum polarization and badly underpredicts the running above ~10 GeV (where muon, tau, and quark loops contribute). The full multi-loop QED calculation would achieve R² > 0.99.
With $n = 1$, the RST correction term becomes $A \cdot \Lambda/Q$ — a hyperbolic $1/Q$ decay. The actual QED running is logarithmic ($\Delta(1/\alpha) \propto \ln(Q/m_e)$). The RST model concentrates all variation near $Q \sim \Lambda$, while the data varies gradually across 5 decades.
![[standard_model_qed.png]]
*QED inverse coupling: RST (n=1, red) vs 1-loop (blue dashed) vs PDG data (black dots).*
---
## 4. Weak (SU(2), n = 3)
### Constrained fit (n = 3)
| Parameter | Value |
|:---|---:|
| $s_0$ | 0.2353 |
| A | 0.000 |
| $\Lambda$ | 340 GeV |
| **R²** | **0.161** |
| RMSE | 0.0032 |
### Interpretation
The optimizer sets A = 0, reducing the model to a constant. The weak mixing angle data is non-monotonic: $\sin^2\theta_W$ rises slightly from 0.2386 (Q = 0) to 0.2397 (Q = 0.16 GeV) before falling to 0.2312 (Z pole). The additive RST model assumes monotonic decrease with $Q$ and cannot capture this turn-around. The variation is also extremely small ($\Delta = 0.007$, or 3% of baseline), making it difficult to distinguish from noise.
The weak sector is the least constrained of the three: fewer data points, non-monotonic running, and a mixing angle rather than a simple coupling constant.
![[standard_model_weak.png]]
*Weak mixing angle: RST (n=3, red) vs PDG data (black dots).*
---
## 5. Discovery mode (free n)
| Force | n (gauge) | n (fitted) | Match % | R² | RMSE |
|:---|---:|---:|---:|---:|---:|
| **QCD** | 8 | 0.32 | 4% | 0.996 | 0.005 |
| **QED** | 1 | 0.15 | 15% | 0.794 | 1.33 |
| **Weak** | 3 | 0.16 | 5% | 0.295 | 0.003 |
All three forces prefer very low $n$ (0.15–0.32), far below the gauge generator counts. This is the optimizer's response to the algebraic-logarithmic mismatch: with very low $n$, the fidelity transition becomes extremely gradual, spanning many decades and roughly mimicking logarithmic behavior.
![[standard_model_discovery.png]]
*Discovery mode: free n vs gauge generator count for all three forces.*
---
## 6. The RST-logarithmic boundary
The results reveal a fundamental boundary, analogous to the RST-Arrhenius boundary in fluids and the RST-Gaussian boundary in dynamical friction:
| Boundary | Classical analog | Quantum occurrence |
|:---|:---|:---|
| **RST-Arrhenius** | Glycerol viscosity | — |
| **RST-Gaussian** | Chandrasekhar friction | — |
| **RST-logarithmic** | — | **Coupling constant running** |
The RST fidelity function $\mu(\eta, n)$ is algebraic: it produces power-law transitions ($\sim Q^{-n}$ at high $Q$). Perturbative QFT coupling running is logarithmic ($\sim 1/\ln(Q/\Lambda)$). These are fundamentally different functional forms:
- **At intermediate Q**: they agree qualitatively (coupling decreases with $Q$ for QCD, increases for QED).
- **At high Q**: RST decays too fast (algebraic tail), while perturbative QCD decays too slowly (logarithmic tail).
- **At low Q**: both diverge, but RST diverges as a power law while QCD diverges as $1/\ln$.
### What this means for the theory
1. **The gauge generator → n mapping is structural**, not phenomenological. It tells us the topological dimensionality of each force's gauge orbit. But the topological prediction (n = dim(g)) describes the *geometry* of the channel, not the *running* of the coupling.
2. **Coupling running is inherently logarithmic** because it arises from loop integrals (vacuum polarization, gluon self-energy) that produce $\ln(Q/\Lambda)$ terms. This logarithmic character is a signature of **quantum** processes — it has no classical analog.
3. **The RST-logarithmic boundary is the Quantum Gate.** It marks the transition from classical physics (where $\mu$ works perfectly) to quantum field theory (where loop corrections introduce transcendental functions). Crossing this boundary requires extending the fidelity function to incorporate logarithmic corrections — perhaps a "quantum fidelity" that includes loop counting.
---
## 7. Physically meaningful findings
Despite the poor quantitative fits, the calibration reveals three physically meaningful results:
1. **The hierarchy is correct.** QCD (n=8) has the sharpest transition, Weak (n=3) is intermediate, QED (n=1) is the most gradual. This matches the physical hierarchy: confinement is catastrophic, the electroweak transition is moderate, and QED running is nearly imperceptible.
2. **The RST Lambda tracks the physical scale.** QCD's optimal $\Lambda = 2545$ GeV places the transition above the data range, consistent with the physical picture that all observed QCD data is in the asymptotic freedom (high-fidelity) regime. QED's $\Lambda = 0.006$ GeV is near $m_e$, the threshold where virtual pairs activate.
3. **n = 0.32 (QCD discovery) recovers the correct shape.** The discovery fit's R² = 0.996 for QCD shows that the additive RST form CAN capture the running shape — but only by abandoning the gauge-derived n. This suggests that a modified model (e.g., $\mu$ composed with a logarithmic map) could reconcile gauge topology with observable running.
---
## 8. Links
- **Theory:** [[Standard Model (RST)]]
- **Code:** [[Standard Model - Code]]
- **Topological spectrum:** [[expanded theory/The Spectrum of Relational Topologies]]
- **Fidelity Inversion:** [[expanded theory/Fidelity Inversion — Gravity and Materials]]
- **Electronic transport (classical analog):** [[../Electronic Transport/Electronic Transport Results]]
- **Fluid dynamics (RST-Arrhenius boundary):** [[../Fluid Dynamics/Fluid Dynamics Results]]
- **Relational dynamics (RST-Gaussian boundary):** [[../Relational Dynamics/Relational Dynamics Results]]