CaloFlow II: Even Faster and Still Accurate Generation of Calorimeter Showers with Normalizing Flows

Krause, Claudius; Shih, David

doi:10.48550/arxiv.2110.11377

Cited by 22 publications

(25 citation statements)

References 22 publications

(55 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, we take a more conservative perspective here. been shown to precisely learn complex distributions in particle physics [33][34][35][36][37][38][39][40][41][42][43]. The statistical benefits of using generative models are discussed in Ref.…”

Section: Online Trainingmentioning

confidence: 99%

Ephemeral Learning -- Augmenting Triggers with Online-Trained Normalizing Flows

Butter,

Diefenbacher,

Kasieczka

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The large data rates at the LHC require an online trigger system to select relevant collisions. Rather than compressing individual events, we propose to compress an entire data set at once. We use a normalizing flow as a deep generative model to learn the probability density of the data online. The events are then represented by the generative neural network and can be inspected offline for anomalies or used for other analysis purposes. We demonstrate our new approach for a toy model and a correlation-enhanced bump hunt.

show abstract

Section: Online Trainingmentioning

confidence: 99%

Ephemeral Learning -- Augmenting Triggers with Online-Trained Normalizing Flows

Butter,

Diefenbacher,

Kasieczka

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Theory-driven ML-generators at the parton level [17,68] can be combined with experimentdriven fast detector simulations [69][70][71][72][73][74][75][76][77][78][79][80] into single generative networks [67,[81][82][83][84][85], provided we have sufficient control over the network and its uncertainties. Single, soup-to-nuts simulation networks are inspired by the fundamental goal of the detection process, namely to reconstruct parton-level information as accurately as possible.…”

Section: Fast Generative Networkmentioning

confidence: 99%

Machine Learning and LHC Event Generation

Butter¹,

Plehn²,

Schumann³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

First-principle simulations are at the heart of the high-energy physics research program. They link the vast data output of multi-purpose detectors with fundamental theory predictions and interpretation. This review illustrates a wide range of applications of modern machine learning to event generation and simulation-based inference, including conceptional developments driven by the specific requirements of particle physics. New ideas and tools developed at the interface of particle physics and machine learning will improve the speed and precision of forward simulations, handle the complexity of collision data, and enhance inference as an inverse simulation problem.

show abstract

“…Inverse Autoregressive Flows (IAFs) [148] are fast in sampling, but a factor d slower in estimating the density of data points. Only recently NFs (based on MAF and IAF architectures) have been applied to calorimeter shower simulation [149,150], surpassing the quality of showers generated by an older GAN trained on the same dataset [122,123].…”

Section: Learning the Simulated Particle Interactions With Matter Thr...mentioning

confidence: 99%

“…The first generative model capable of generating samples that would confuse a classifier were based on normalizing flows [149,150]. This approach, called CaloFlow, used the same detector geometry as the CaloGAN.…”

Section: Hybrid Calorimeter For a Future Particle Collidermentioning

confidence: 99%

Toward the End-to-End Optimization of Particle Physics Instruments with Differentiable Programming: a White Paper

Dorigo¹,

Giammanco²,

Vischia³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The full optimization of the design and operation of instruments whose functioning relies on the interaction of radiation with matter is a superhuman task, given the large dimensionality of the space of possible choices for geometry, detection technology, materials, data-acquisition, and informationextraction techniques, and the interdependence of the related parameters. On the other hand, massive potential gains in performance over standard, "experience-driven" layouts are in principle within our reach if an objective function fully aligned with the final goals of the instrument is maximized by means of a systematic search of the configuration space. The stochastic nature of the involved quantum processes make the modeling of these systems an intractable problem from a classical statistics point of view, yet the construction of a fully differentiable pipeline and the use of deep learning techniques may allow the simultaneous optimization of all design parameters.In this document we lay down our plans for the design of a modular and versatile modeling tool for the end-to-end optimization of complex instruments for particle physics experiments as well as industrial and medical applications that share the detection of radiation as their basic ingredient. We consider a selected set of use cases to highlight the specific needs of different applications.

show abstract

CaloFlow II: Even Faster and Still Accurate Generation of Calorimeter Showers with Normalizing Flows

Cited by 22 publications

References 22 publications

Ephemeral Learning -- Augmenting Triggers with Online-Trained Normalizing Flows

Ephemeral Learning -- Augmenting Triggers with Online-Trained Normalizing Flows

Machine Learning and LHC Event Generation

Toward the End-to-End Optimization of Particle Physics Instruments with Differentiable Programming: a White Paper

Contact Info

Product

Resources

About