Machine learning speeds BSM data interpretation at LHC.
problem Challenging to scan high-dimensional BSM theories due to expensive simulations.
method Machine learning to predict BSM theory parameters from data.
result Predicts natural SUSY events up to 4 orders of magnitude faster.
Tensor networks improve anomaly detection at LHC for new physics.
problem Identifying new phenomena in proton collision events at LHC.
method Tensor network-based anomaly detection using Matrix Product State with an isometric feature map.
result Tensor networks outperform established quantum methods in identifying new phenomena.
DVAEs speed up calorimeter simulation for LHC data.
problem Slow calorimeter simulation in LHC experiments.
method Discrete Variational Autoencoders (DVAEs).
result Significantly faster calorimeter shower simulation.
New methods combine matrix elements and machine learning for more accurate LHC measurements.
problem High-dimensional data and complex detector response make likelihood function estimation difficult.
method Review and application of traditional histogram, Matrix Element Method, Optimal Observables, and neural density estimation techniques. Use of MadMiner for automation.
result New techniques have the potential to substantially improve LHC measurement sensitivity.
Challenge uses unsupervised learning to detect new physics signals at LHC.
problem Detecting new physics signals at the LHC using unsupervised machine learning.
method Developed and evaluated anomaly detection algorithms on a large dataset.
result Benchmark dataset of >1 Billion simulated LHC events for future studies.
New methods improve precision of LHC measurements.
problem Inference on high-dimensional LHC data is difficult due to complex simulators.
method Simulation-based inference methods combining machine learning and simulator information.
result These techniques have the potential to substantially improve LHC measurements.
New algorithm for high-dimensional LHC measurements using deep learning.
problem Efficiency correction in high-dimensional LHC measurements.
method Probabilistic Deep Neural Network trained on detector simulation and phase space events.
result Optimal fiducial measurements independent of event generators.
PHAZE framework uses zkML and hashing for fast, verifiable LHC trigger decisions.
problem Inefficient inference on large machine learning models for LHC trigger performance.
method Cryptographic techniques like hashing and zkML for low latency, certifiable inference.
result Achieves nanosecond-order latency for LHC triggers, enabling dynamic low-level triggers.
Graph neural networks improve charged particle tracking on FPGAs.
problem Charged particle trajectory determination in high interaction density conditions.
method Graph neural networks (GNNs) embedded in tracker data as graphs, classifying edges as track segments.
result GNNs implemented on FPGAs for charged particle tracking, enabling future HL-LHC experiments.
Generative model speeds up particle shower simulations in calorimeters.
problem Expensive and slow particle collision simulations for LHC experiments.
method Deep neural network for high-fidelity, fast electromagnetic calorimeter simulation.
result Achieves speed-up factors of up to 100,000x while maintaining accuracy.
New methods use machine learning to tighten constraints on particle physics theories.
problem Improving precision of LHC legacy constraints on particle physics theories.
method Machine learning techniques applied to Monte-Carlo simulations of particle physics processes.
result Significantly stronger bounds on dimension-six operators achieved.
Study compares unsupervised and weakly-supervised methods for anomaly detection at the LHC.
problem Detecting new physics signals at the LHC with model-agnostic techniques.
method Compared unsupervised autoencoder (AE) and weakly-supervised Classification Without Labels (CWoLa) methods.
result Both methods complement each other, providing sensitivity to different types of signals.
Study interprets deep learning for LHC jet tagging.
problem Understanding deep learning models in LHC jet tagging.
method Recursive neural networks, comparative study of jet tagging tasks.
result Interesting observations on the latent space of jet tagging models.
Enhanced detection of sneutrinos at the LHC using machine learning.
problem Detecting rare new physics signals in the presence of significant backgrounds.
method Machine learning models (XGBoost and deep neural network) applied to template fit analysis.
result Template fit outperforms simple cuts in enhancing sneutrino detectability.
The Inverse Bagging Algorithm detects anomalies by identifying sub-samples rich in known data.
problem Detecting anomalies in data sets with a well-modeled process and an unknown PDF.
method Uses inverse bootstrap aggregating to identify sub-samples rich in the known process and classify events.
result The method avoids modifying the kinematic distributions of the well-modeled process.
DeepJet improves jet flavor classification and quark-gluon tagging.
problem Jet flavor classification in high-energy physics experiments.
method Proposes a novel deep learning architecture, DeepJet, for improved performance.
result Improves heavy flavor classification performance and extends to quark-gluon tagging.
LSTM networks improve top jet tagging at the LHC.
problem Boosted top quark tagging at the LHC.
method Incorporating Long Short-Term Memory (LSTM) networks into jet constituent analysis.
result Best LSTM network achieves 100 background rejection at 50% signal efficiency.
ALICE uses ML for particle identification across a wide momentum range.
problem Effective combination of detector information for particle identification.
method Machine Learning, specifically Random Forest and Domain Adaptation Neural Networks.
result Advanced ML solutions improve particle identification accuracy.
Two methods use simulation to improve anomaly detection in particle physics.
problem Artificial bumps in invariant mass spectra from machine learning classifiers.
method Simulation-assisted decorrelation techniques.
result Both methods are robust to correlations in the data and improve anomaly detection.
ANODE uses neural density estimation for anomaly detection in physics.
problem Detecting localized anomalies in signal regions with limited background information.
method Estimate data and background densities, construct likelihood ratio, and enhance significance.
result ANODE enhances dijet bump hunt significance by up to 7x with 10% background accuracy.
Graph networks improve particle reconstruction in irregular detectors.
problem Handling irregular particle-detector geometries in particle reconstruction.
method Introduce distance-weighted graph network architectures (GarNet, GravNet layers) for irregular geometry detectors.
result The proposed graph networks provide equally performing or less resource-demanding solutions compared to existing methods.
New method uses machine learning to estimate sensitivity without binning.
problem Estimating sensitivity of high-dimensional data sets without binning.
method Combines machine-learning classification with likelihood-based inference tests using Kernel Density Estimators.
result Significance estimation is not sensitive to non-smooth probability distributions.
MLPF uses graph neural networks to improve particle-flow reconstruction in high-pileup conditions.
problem Improving particle-flow reconstruction in high-pileup conditions at high-luminosity LHC.
method End-to-end trainable machine-learned particle-flow algorithm based on graph neural networks.
result MLPF improves physics response and demonstrates scalable reconstruction in high-pileup environments.
New method uses cluster shapes to improve track finding in particle collisions.
problem Combining timing and additional detector information for efficient track finding.
method Neural networks to analyze cluster shapes for track seeding.
result Cluster shapes reduce fake combinatorial backgrounds while maintaining high track efficiency.
A new jet constituent-based method for top quark tagging achieves high background rejection.
problem Tagging highly energetic jets resulting from top quark decays.
method Sequential approach using ordered jet constituents as inputs, avoiding loss of information.
result Achieves a background rejection of 45 at a 50% efficiency operating point.
New method uses less data to train machine learning models.
problem Maximizing machine learning algorithm performance with limited data.
method Compare weakly supervised (class ratios) vs fully supervised (truth-level labels) methods.
result Weakly supervised networks yield impressive results with less data.
Quantum GNNs outperform classical GNNs in jet tagging.
problem Classifying partons initiating jets from high-energy particle collisions.
method Comparison of classical and quantum GNNs and their equivariant counterparts.
result Quantum GNNs outperformed classical GNNs in binary classification tasks.
FPGAs enable real-time neural network inference for particle physics.
problem Low-latency, low-power requirements for particle physics.
method Developed hls4ml for building machine learning models in FPGAs.
result Neural network inference fits within modern FPGA resources with 100 ns latency.
Imbalanced data sets containing much more background than signal instances are very common in particle physics, and will also be characteristic for the upcoming analyses of LHC data. Following up the work presented at ACAT 2008, we use the multivariate technique presented there (a rule growing algorithm with the meta-m…
VAE improves anomaly detection for jet tagging at the LHC.
problem Anomaly detection in jet tagging at the LHC.
method Variational Autoencoder (VAE) trained on background QCD jets, with latent space learning for anomaly detection.
result Outlier Exposed VAE (OE-VAE) achieves excellent results in both sensitivity and decorrelation of jet mass.
CMS uses neural networks to monitor muon detector anomalies.
problem Monitoring anomalies in muon detector data for physics analysis.
method Supervised and semi-supervised artificial neural networks, convolutional autoencoders.
result Unprecedented efficiency in detecting known anomalous behaviors.
Foundation models trained on collider data improve jet generation tasks.
problem Improving foundation models for jet generation tasks.
method Pre-training OmniJet-α model on AspenOpenJets dataset. result Pre-trained model improves performance on jet generation tasks with domain shift.
MadMiner uses machine learning to analyze high-dimensional particle physics data.
problem Challenges in analyzing high-dimensional event data for subtle kinematic signatures.
method Combines matrix element information and machine learning to analyze particle physics data.
result New techniques substantially increase sensitivity to new physics.
Collimated streams of particles produced in high energy physics experiments are organized using clustering algorithms to form jets. To construct jets, the experimental collaborations based at the Large Hadron Collider (LHC) primarily use agglomerative hierarchical clustering schemes known as sequential recombination. W…
Iterative subtraction method outperforms other feature ranking techniques in high-energy physics.
problem Determining the most important features for classification in high-energy physics experiments.
method Comparison of feature ranking methods including Iterative Addition, Iterative Removal, and BDT Selection Frequency.
result Iterative Removal method is the most efficient for feature ranking in classification tasks.
HI-SIGMA improves sensitivity in high-dimensional statistical inference with data-driven background models.
problem Performing high-dimensional statistical inference with complex backgrounds in high-energy physics.
method HI-SIGMA uses generative ML models to learn signal and background distributions, incorporating systematic uncertainties.
result HI-SIGMA provides improved sensitivity compared to classifier-based methods.
End-to-end Sinkhorn Autoencoder reduces data simulation time with noise generation.
problem Efficiently simulating data collection processes with high fidelity and speed.
method End-to-end Sinkhorn Autoencoder with noise generator.
result Outperforms competing methods on various datasets.
Develops machine learning classifiers for better centrality estimation in proton-nucleus and nucleus-nucleus collisions.
problem Direct measurement of centrality in A-A and p-A collisions is challenging due to limited data access.
method Uses machine learning techniques to classify centrality based on information from multiple detector subsystems.
result Improved centrality resolution can reduce volume fluctuations impact on physical observables.
New method combines simulations and data for anomaly detection.
problem Detecting new particle signals without direct evidence.
method Hybrid approach using reweighting and interpolation.
result Improved background estimation and classification.
Recursive neural networks mimic QCD for jet physics.
problem Improving jet physics predictions using machine learning.
method Analogies between QCD and natural languages for jet clustering.
result Recursive architectures are more accurate and data efficient than previous methods.
Adversarial domain adaptation reduces sample bias in high energy physics classifier.
problem Sample bias in high energy physics classifier training.
method Adversarial domain adaptation using neural networks with gradient reversal layer.
result Successful bias removal on simulated events at the LHC.
Improved sensitivity to Higgs potential through neural simulation-based inference for di-Higgs events.
problem Improving sensitivity to physics beyond the Standard Model through di-Higgs events.
method Simulation-based inference using neural networks to estimate per-event likelihood ratios.
result Adding kinematic observables improves experimental sensitivity to Higgs self-coupling.
New methods use machine learning to constrain particle physics parameters.
problem Constraining effective field theory parameters in collider experiments.
method Inference techniques using latent-space structure and neural networks trained on augmented data.
result Significantly stronger bounds on effective dimension-six operators.
Boosted decision trees improved for particle identification in high-energy physics.
problem Overfitting in boosted decision trees hampers their performance in particle identification.
method Meta-learning techniques of boosting and bagging to mitigate overfitting.
result The proposed algorithm achieves performance close to that of deep neural networks on a benchmark data set.
Optimal transport calibrates machine learning models for particle physics simulations.
problem Discrepancies between simulation and experimental data limit machine learning effectiveness.
method A model calibration approach based on optimal transport applied to high-dimensional simulations.
result Calibrated high-dimensional representations enable proper calibration of various downstream quantities.
Etalumis bridges scientific simulators and probabilistic programming.
problem Infeasibility of rewriting scientific simulators for Bayesian inference.
method Cross-platform probabilistic execution protocol, MCMC and IC engines, distributed training of 3DCNN-LSTM.
result Achieved largest-scale posterior inference in a Turing-complete PPL for LHC use-case.
Proposes a meta-algorithm for classification with overlapping classes in high-energy physics.
problem Challenges of class overlap in binary classification.
method Combines bagging and boosting techniques with a randomization trick.
result Improves statistical significance of Higgs discovery.
CaloGAN speeds up particle shower simulations in calorimeters.
problem Accurate simulation of particle showers in calorimeters for high-energy physics experiments.
method Generative adversarial networks (GANs) for fast simulation.
result Achieved speedup factors of up to 100,000x compared to full simulation techniques.