Search Results for author: Gregor Kasieczka

Found 24 papers, 11 papers with code

Convolutional L2LFlows: Generating Accurate Showers in Highly Granular Calorimeters Using Convolutional Normalizing Flows

1 code implementation30 May 2024 Thorsten Buss, Frank Gaede, Gregor Kasieczka, Claudius Krause, David Shih

In the quest to build generative surrogate models as computationally efficient alternatives to rule-based simulations, the quality of the generated samples remains a crucial frontier.

OmniJet-$α$: The first cross-task foundation model for particle physics

no code implementations8 Mar 2024 Joschka Birk, Anna Hallin, Gregor Kasieczka

This is the first successful transfer between two different and actively studied classes of tasks and constitutes a major step in the building of foundation models for particle physics.

Jet Tagging Transfer Learning

Sets are all you need: Ultrafast jet classification on FPGAs for HL-LHC

no code implementations2 Feb 2024 Patrick Odagiu, Zhiqiang Que, Javier Duarte, Johannes Haller, Gregor Kasieczka, Artur Lobanov, Vladimir Loncar, Wayne Luk, Jennifer Ngadiuba, Maurizio Pierini, Philipp Rincke, Arpita Seksaria, Sioni Summers, Andre Sznajder, Alexander Tapper, Thea K. Aarrestad

We study various machine learning based algorithms for performing accurate jet flavor classification on field-programmable gate arrays and demonstrate how latency and resource consumption scale with the input size and choice of algorithm.

Quantization

AdamMCMC: Combining Metropolis Adjusted Langevin with Momentum-based Optimization

no code implementations21 Dec 2023 Sebastian Bieringer, Gregor Kasieczka, Maximilian F. Steffen, Mathias Trabs

Uncertainty estimation is a key issue when considering the application of deep neural network methods in science and engineering.

Residual ANODE

no code implementations18 Dec 2023 Ranit Das, Gregor Kasieczka, David Shih

We present R-ANODE, a new method for data-driven, model-agnostic resonant anomaly detection that raises the bar for both performance and interpretability.

Anomaly Detection Inductive Bias

Statistical guarantees for stochastic Metropolis-Hastings

1 code implementation13 Oct 2023 Sebastian Bieringer, Gregor Kasieczka, Maximilian F. Steffen, Mathias Trabs

A Metropolis-Hastings step is widely used for gradient-based Markov chain Monte Carlo methods in uncertainty quantification.

regression Uncertainty Quantification

EPiC-ly Fast Particle Cloud Generation with Flow-Matching and Diffusion

no code implementations29 Sep 2023 Erik Buhmann, Cedric Ewen, Darius A. Faroughy, Tobias Golling, Gregor Kasieczka, Matthew Leigh, Guillaume Quétant, John Andrew Raine, Debajyoti Sengupta, David Shih

In addition, we introduce \epcfm, the first permutation equivariant continuous normalizing flow (CNF) for particle cloud generation.

CaloClouds II: Ultra-Fast Geometry-Independent Highly-Granular Calorimeter Simulation

1 code implementation11 Sep 2023 Erik Buhmann, Frank Gaede, Gregor Kasieczka, Anatolii Korol, William Korcari, Katja Krüger, Peter McKeown

We further distill the diffusion model into a consistency model allowing for accurate sampling in a single step and resulting in a $46\times$ ($37\times$ over CaloClouds) speed-up.

CaloClouds: Fast Geometry-Independent Highly-Granular Calorimeter Simulation

2 code implementations8 May 2023 Erik Buhmann, Sascha Diefenbacher, Engin Eren, Frank Gaede, Gregor Kasieczka, Anatolii Korol, William Korcari, Katja Krüger, Peter McKeown

Simulating showers of particles in highly-granular detectors is a key frontier in the application of machine learning to particle physics.

EPiC-GAN: Equivariant Point Cloud Generation for Particle Jets

1 code implementation17 Jan 2023 Erik Buhmann, Gregor Kasieczka, Jesse Thaler

With the vast data-collecting capabilities of current and future high-energy collider experiments, there is an increasing demand for computationally efficient simulations.

Generative Adversarial Network Point Cloud Generation

Feature Selection with Distance Correlation

no code implementations30 Nov 2022 Ranit Das, Gregor Kasieczka, David Shih

Choosing which properties of the data to use as input to multivariate decision algorithms -- a. k. a.

Automated Feature Engineering Feature Engineering +1

Data Science and Machine Learning in Education

no code implementations19 Jul 2022 Gabriele Benelli, Thomas Y. Chen, Javier Duarte, Matthew Feickert, Matthew Graham, Lindsey Gray, Dan Hackett, Phil Harris, Shih-Chieh Hsu, Gregor Kasieczka, Elham E. Khoda, Matthias Komm, Mia Liu, Mark S. Neubauer, Scarlet Norberg, Alexx Perloff, Marcel Rieger, Claire Savard, Kazuhiro Terao, Savannah Thais, Avik Roy, Jean-Roch Vlimant, Grigorios Chachamis

The growing role of data science (DS) and machine learning (ML) in high-energy physics (HEP) is well established and pertinent given the complex detectors, large data, sets and sophisticated analyses at the heart of HEP research.

BIG-bench Machine Learning

Machine Learning in the Search for New Fundamental Physics

no code implementations7 Dec 2021 Georgia Karagiorgi, Gregor Kasieczka, Scott Kravitz, Benjamin Nachman, David Shih

Machine learning plays a crucial role in enhancing and accelerating the search for new fundamental physics.

BIG-bench Machine Learning

New Methods and Datasets for Group Anomaly Detection From Fundamental Physics

no code implementations6 Jul 2021 Gregor Kasieczka, Benjamin Nachman, David Shih

The identification of anomalous overdensities in data - group or collective anomaly detection - is a rich problem with a large number of real world applications.

Group Anomaly Detection

Shared Data and Algorithms for Deep Learning in Fundamental Physics

1 code implementation1 Jul 2021 Lisa Benato, Erik Buhmann, Martin Erdmann, Peter Fackeldey, Jonas Glombitza, Nikolai Hartmann, Gregor Kasieczka, William Korcari, Thomas Kuhr, Jan Steinheimer, Horst Stöcker, Tilman Plehn, Kai Zhou

We introduce a Python package that provides simply and unified access to a collection of datasets from fundamental physics research - including particle physics, astroparticle physics, and hadron- and nuclear physics - for supervised machine learning studies.

BIG-bench Machine Learning Transfer Learning

DCTRGAN: Improving the Precision of Generative Models with Reweighting

2 code implementations3 Sep 2020 Sascha Diefenbacher, Engin Eren, Gregor Kasieczka, Anatolii Korol, Benjamin Nachman, David Shih

We introduce a post-hoc correction to deep generative models to further improve their fidelity, based on the Deep neural networks using the Classification for Tuning and Reweighting (DCTR) protocol.

GANplifying Event Samples

no code implementations14 Aug 2020 Anja Butter, Sascha Diefenbacher, Gregor Kasieczka, Benjamin Nachman, Tilman Plehn

A critical question concerning generative networks applied to event generation in particle physics is if the generated events add statistical precision beyond the training sample.

Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed

2 code implementations11 May 2020 Erik Buhmann, Sascha Diefenbacher, Engin Eren, Frank Gaede, Gregor Kasieczka, Anatolii Korol, Katja Krüger

Accurate simulation of physical processes is crucial for the success of modern particle physics.

Instrumentation and Detectors High Energy Physics - Experiment High Energy Physics - Phenomenology Data Analysis, Statistics and Probability

DisCo Fever: Robust Networks Through Distance Correlation

1 code implementation13 Jan 2020 Gregor Kasieczka, David Shih

While deep learning has proven to be extremely successful at supervised classification tasks at the LHC and beyond, for practical applications, raw classification accuracy is often not the only consideration.

High Energy Physics - Phenomenology High Energy Physics - Experiment Data Analysis, Statistics and Probability

Deep-learning Top Taggers or The End of QCD?

1 code implementation30 Jan 2017 Gregor Kasieczka, Tilman Plehn, Michael Russell, Torben Schell

Machine learning based on convolutional neural networks can be used to study jet images from the LHC.

High Energy Physics - Phenomenology

Resonance Searches with an Updated Top Tagger

no code implementations19 Mar 2015 Gregor Kasieczka, Tilman Plehn, Torben Schell, Thomas Strebler, Gavin P. Salam

The performance of top taggers, for example in resonance searches, can be significantly enhanced through an increased set of variables, with a special focus on final-state radiation.

High Energy Physics - Phenomenology

Cannot find the paper you are looking for? You can Submit a new open access paper.