The Neyman–Pearson theory of hypothesis testing can be employed for goodness of fit if the alternative hypothesis H1 is generic enough not to introduce a significant bias while at the same time avoiding overfitting. A practical implementation of this idea (dubbed NPLM) has been recently developed in the context of high energy physics to determine the compatibility of data with a reference model. After a general introduction to the approach, I will present a specific algorithm based on efficient machine learning methods and discuss two applications: model-independent new physics searches and data quality monitoring.
References:
https://arxiv.org/abs/2204.02317
https://arxiv.org/abs/2303.05413
Valerio Ippolito