Speaker
Description
As generative models start taking an increasingly prominent role in both particle physics and everyday life, quantifying the statistical power and expressiveness of such generative models becomes a more and more pressing question.
In past work, we have seen that a generative mode can, in fact, be used to generate samples beyond the initial training data. However, the exact quantification of the amplification factor between the statistical power of original training data and the statistical power of generated samples has had to rely on knowledge of the true distribution in some form.
We present a new approach of the prediction of the amplification factor of a generative model, which does not require knowledge of the true distribution, and demonstrate this both on meaningful constructed examples and on relevant physics datasets.
Significance
Generative amplification of datasets forms the underpinning principle of nearly all work on generative fast simulation. Being able to quantify the exact amplification factor without the need of a vastly larger reference dataset presents a vital step in the general application of fast simulation methods and generative approach in particle physics.
