Quantum mechanics

Linguistics

Philosophy

Bayesian probability

Pure mathematics

Bayesian network

Bayes' theorem

Event (particle physics)

Analogy

Probabilistic logic

Computer science

Physics

Generative grammar

Mathematics

Artificial intelligence

Invertible matrix

Artificial neural network

Machine learning

SciPost Physics

Following the growing success of generative neural networks in LHC simulations, the crucial question is how to control the networks and assign uncertainties to their event output. We show how Bayesian normalizing flows or invertible networks capture uncertainties from the training and turn them into an uncertainty on the event weight. Fundamentally, the interplay between density and uncertainty estimates indicates that these networks learn functions in analogy to parameter fits rather than binned event counts.

Understanding Event-Generation Networks via Uncertainties