Speaker
Description
In the contemporary landscape of advanced statistical analysis toolkits, ranging from Bayesian inference to machine learning, the seemingly straightforward concept of a histogram often goes unnoticed. However, the power and compactness of partially aggregated, multi-dimensional summary statistics with a fundamental connection to differential and integral calculus make them formidable statistical objects. Expressing these concepts robustly and efficiently in high-dimensional parameter spaces is a non-trivial challenge, especially when the resulting library is meant to remain usable by scientists rather than software engineers.
A decade after its initial release, the YODA statistical library has been redesigned from the ground, aiming to generalise its principles while addressing real-world usage requirements in the era of expanding computational power and vast datasets. We will summarise the core principles required for consistent generalised histogramming and outline some of the C++ metaprogramming techniques adopted to handle dimensionality relationships in the revamped YODA histogramming library. Used both in Rivet and Contur, YODA is a key component of physics data–model comparison and statistical interpretation in collider physics.
References
https://arxiv.org/abs/2312.15070
Significance
The YODA library is a key component in the Rivet and Contur packages. 10 years after its initial release, YODA has been redesigned using modern C++ techniques to provide generalised histogramming in arbitrary dimensions and addressing various other shortcomings of the initial release series.
