README.md

Fits to SMEFT coefficients II and Machine Learning

This series of notebooks contain all the code to reproduce the data in Chapter 7 of my PhD thesis, which is adapted from 2109.07405

• 01: Fits to Scenarios I and II and their errors. Pulls in Scenario II. Leptoquark couplings.
• 02: Plots for $\alpha^l$ vs $\beta^\ell$ and $\alpha^q$ vs $\beta^q$ in Scenario II.
• 03: Plots for $R_{K^{()}}$ and $R_{D^{()}}$ observables in Scenario I and II.
• 04: Preparation of the datasets to train the Machine Learning algorithm.
• 05: Training of the ML algorithm. Regression, histogram and SHAP values.
• 06: Correlation matrices for Wilson Coefficients and for observables in the dataset generated by the ML algorithm.

Additional files

• Hessian ellipse data for:
  • Scenario I
  • Scenario II
• Likelihood plots for:
  • $\alpha^\ell$ and $\beta^\ell$: (pdf) | (pgf) | (dat)
  • $\alpha^q$ and $\beta^q$: (pdf) | (pgf) | (dat)
• $R_{K^{()}}$ and $R_{D^{()}}$ observables:
  • Scenario I
  • Scenario II
• Confidence plots for observables:
  • $R_{K^{(*)}}$: (pdf) | (pgf)
  • $R_{D^{(*)}}$: (pdf) | (pgf)
• Plot of the pulls for every observable in Scenario II: (pdf) | (pgf)
• List of the pulls for every observable in Scenario II: (pdf) | (tex)
• Evolution of the likelihood changing one parameter:
  • $C$: (pdf) | (pgf)
  • $\alpha^\ell$: (pdf) | (pgf)
  • $\beta^\ell$: (pdf) | (pgf)
  • $\alpha^q$: (pdf) | (pgf)
  • $\beta^q$: (pdf) | (pgf)
• Regression plot for the training of the Machine Learning algorithm
• Histogram of the ML-generated points: (pdf) | (pgf)
• SHAP summary plot
• SHAP values for the fit parameters:
  • $C$
  • $\alpha^\ell$
  • $\beta^\ell$
  • $\alpha^q$
  • $\beta^q$
• Correlation matrices:
  • Between WET Wilson coefficients
  • Between observables
• Plot for $R_D$ and $\mathrm{B\to K\nu\nu}$