Detecting Model Misspecification in Amortized Bayesian Inference with Neural Networks: An Extended Investigation
Recent advances in probabilistic deep learning enable efficient amortized Bayesian
inference in settings where the likelihood function is only implicitly defined by a simulation …
inference in settings where the likelihood function is only implicitly defined by a simulation …
Differentiable MadNIS-Lite
T Heimel, O Mattelaer, T Plehn… - arXiv preprint arXiv …, 2024 - arxiv.org
Differentiable programming opens exciting new avenues in particle physics, also affecting
future event generators. These new techniques boost the performance of current and …
future event generators. These new techniques boost the performance of current and …
CaloDREAM $\unicode {x2013} $ Detector Response Emulation via Attentive flow Matching
Detector simulations are an exciting application of modern generative networks. Their
sparse high-dimensional data combined with the required precision poses a serious …
sparse high-dimensional data combined with the required precision poses a serious …
Optimal, fast, and robust inference of reionization-era cosmology with the 21cmPIE-INN
Modern machine learning will allow for simulation-based inference from reionization-era
21cm observations at the Square Kilometre Array. Our framework combines a convolutional …
21cm observations at the Square Kilometre Array. Our framework combines a convolutional …
Implicit Quantile Networks For Emulation in Jet Physics
B Kronheim, AA Kadhim, MP Kuchera… - arXiv preprint arXiv …, 2023 - arxiv.org
The ability to model and sample from conditional densities is important in many physics
applications. Implicit quantile networks (IQN) have been successfully applied to this task in …
applications. Implicit quantile networks (IQN) have been successfully applied to this task in …
Invertible Neural Networks in Astrophysics
RS Klessen - EPJ Web of Conferences, 2022 - epj-conferences.org
Modern machine learning techniques have become indispensable in many fields of
astronomy and astrophysics. Here we introduce a specific class of methods, invertible neural …
astronomy and astrophysics. Here we introduce a specific class of methods, invertible neural …
Event generators for high-energy physics experiments
We provide an overview of the status of Monte-Carlo event generators for high-energy
particle physics. Guided by the experimental needs and requirements, we highlight areas of …
particle physics. Guided by the experimental needs and requirements, we highlight areas of …
Making the most of LHC data-Bayesian neural networks and SMEFT global analysis
MA Luchmann - 2022 - archiv.ub.uni-heidelberg.de
The LHC produces huge amounts of data in which signs of new physics can be hidden. To
take full advantage of existing or future LHC data, it is worth exploring novel techniques such …
take full advantage of existing or future LHC data, it is worth exploring novel techniques such …
arXiv: Normalizing Flows for High-Dimensional Detector Simulations
F Ernst, L Favaro, T Plehn, D Shih, C Krause - 2023 - cds.cern.ch
Whenever invertible generative networks are needed for LHC physics, normalizing flows
show excellent performance. A challenge is their scaling to high-dimensional phase spaces …
show excellent performance. A challenge is their scaling to high-dimensional phase spaces …
The Flow of LHC Events-Generative models for LHC simulations and inference
T Heimel - 2024 - archiv.ub.uni-heidelberg.de
Generative neural networks have various applications in LHC physics, for both fast
simulations and precise inference. We first show that normalizing flows can be used to …
simulations and precise inference. We first show that normalizing flows can be used to …