Sergei V Gleyzer

Machine learning is an important applied research area in particle physics, beginning with applications to high-level physics analysis in the 1990s and 2000s, followed by an explosion of applications in particle and event identification and reconstruction in the 2010s. In this document we discuss promising future research and development areas in machine learning in particle physics with a roadmap for their implementation, software and hardware resource requirements, collaborative initiatives with the data science community, academia and industry, and training the particle physics community in data science. The main objective of the document is to connect and motivate these areas of research and development with the physics drivers of the High-Luminosity Large Hadron Collider and future neutrino experiments and identify the resource needs for their implementation. Additionally we identify areas where collaboration with external communities will be of great benefit.

Abstract Accurate observation of two or more particles within a very narrow time window has always been a challenge in modern physics. It creates the possibility of correlation experiments, such as the ground-breaking Hanbury Brown–Twiss experiment, leading to new physical insights. For low-energy electrons, one possibility is to use a Microchannel plate with subsequent delay lines for the readout of the incident particle hits, a setup called a Delay Line Detector. The spatial and temporal coordinates of more than one particle can be fully reconstructed outside a region called the dead radius. For interesting events, where two electrons are close in space and time, the determination of the individual positions of the electrons requires elaborate peak finding algorithms. While classical methods work well with single particle hits, they fail to identify and reconstruct events caused by multiple nearby particles. To address this challenge, we present a new spatiotemporal machine learning model to identify and reconstruct the position and time of such multi-hit particle signals. This model achieves a much better resolution for nearby particle hits compared to the classical approach, removing some of the artifacts and reducing the dead radius a factor of eight. We show that machine learning models can be effective in improving the spatiotemporal performance of delay line detectors.

Abstract Strong gravitational lensing is a promising probe of the substructure of dark matter halos. Deep-learning methods have the potential to accurately identify images containing substructure, and differentiate weakly interacting massive particle dark matter from other well motivated models, including vortex substructure of dark matter condensates and superfluids. This is crucial in future efforts to identify the true nature of dark matter. We implement, for the first time, a classification approach to identifying dark matter based on simulated strong lensing images with different substructure. Utilizing convolutional neural networks trained on sets of simulated images, we demonstrate the feasibility of deep neural networks to reliably distinguish among different types of dark matter substructure. With thousands of strong lensing images anticipated with the coming launch of Vera C. Rubin Observatory, we expect that supervised and unsupervised deep-learning models will play a crucial role in determining the nature of dark matter.

We describe the construction of end-to-end jet image classifiers based on simulated low-level detector data to discriminate quark- vs. gluon-initiated jets with high-fidelity simulated CMS Open Data. We highlight the importance of precise spatial information and demonstrate competitive performance to existing state-of-the-art jet classifiers. We further generalize the end-to-end approach to event-level classification of quark vs. gluon di-jet QCD events. We compare the fully end-to-end approach to using hand-engineered features and demonstrate that the end-to-end algorithm is robust against the effects of underlying event and pile-up.

The cross section is one of the most important physical quantities in high-energy physics and the most time consuming to compute. While machine learning has proven to be highly successful in numerical calculations in high-energy physics, analytical calculations using machine learning are still in their infancy. In this work, we use a sequence-to-sequence model, specifically, a transformer, to compute a key element of the cross section calculation, namely, the squared amplitude of an interaction. We show that a transformer model is able to predict correctly 97.6% and 99% of squared amplitudes of QCD and QED processes, respectively, at a speed that is up to orders of magnitude faster than current symbolic computation frameworks. We discuss the performance of the current model, its limitations and possible future directions for this work.

Models based on vision transformer architectures are considered state-of-the-art when it comes to image classification tasks. However, they require extensive computational resources both for training and deployment. The problem is exacerbated as the amount and complexity of the data increases. Quantum-based vision transformer models could potentially alleviate this issue by reducing the training and operating time while maintaining the same predictive power. Although current quantum computers are not yet able to perform high-dimensional tasks yet, they do offer one of the most efficient solutions for the future. In this work, we construct several variations of a quantum hybrid vision transformer for a classification problem in high energy physics (distinguishing photons and electrons in the electromagnetic calorimeter). We test them against classical vision transformer architectures. Our findings indicate that the hybrid models can achieve comparable performance to their classical analogues with a similar number of parameters.

Machine learning algorithms are heavily relied on to understand the vast amounts of data from high-energy particle collisions at the CERN Large Hadron Collider (LHC). The data from such collision events can naturally be represented with graph structures. Therefore, deep geometric methods, such as graph neural networks (GNNs), have been leveraged for various data analysis tasks in high-energy physics. One typical task is jet tagging, where jets are viewed as point clouds with distinct features and edge connections between their constituent particles. The increasing size and complexity of the LHC particle datasets, as well as the computational models used for their analysis, have greatly motivated the development of alternative fast and efficient computational paradigms such as quantum computation. In addition, to enhance the validity and robustness of deep networks, we can leverage the fundamental symmetries present in the data through the use of invariant inputs and equivariant layers. In this paper, we provide a fair and comprehensive comparison of classical graph neural networks (GNNs) and equivariant graph neural networks (EGNNs) and their quantum counterparts: quantum graph neural networks (QGNNs) and equivariant quantum graph neural networks (EQGNN). The four architectures were benchmarked on a binary classification task to classify the parton-level particle initiating the jet. Based on their area under the curve (AUC) scores, the quantum networks were found to outperform the classical networks. However, seeing the computational advantage of quantum networks in practice may have to wait for the further development of quantum technology and its associated application programming interfaces (APIs).

Models based on vision transformer architectures are considered state-of-the-art when it comes to image classification tasks. However, they require extensive computational resources both for training and deployment. The problem is exacerbated as the amount and complexity of the data increases. Quantum-based vision transformer models could potentially alleviate this issue by reducing the training and operating time while maintaining the same predictive power. Although current quantum computers are not yet able to perform high-dimensional tasks, they do offer one of the most efficient solutions for the future. In this work, we construct several variations of a quantum hybrid vision transformer for a classification problem in high-energy physics (distinguishing photons and electrons in the electromagnetic calorimeter). We test them against classical vision transformer architectures. Our findings indicate that the hybrid models can achieve comparable performance to their classical analogs with a similar number of parameters.

This paper presents a comparative analysis of the performance of Equivariant Quantum Neural Networks (EQNNs) and Quantum Neural Networks (QNNs), juxtaposed against their classical counterparts: Equivariant Neural Networks (ENNs) and Deep Neural Networks (DNNs). We evaluate the performance of each network with three two-dimensional toy examples for a binary classification task, focusing on model complexity (measured by the number of parameters) and the size of the training dataset. Our results show that the Z2×Z2 EQNN and the QNN provide superior performance for smaller parameter sets and modest training data samples.

Deep learning techniques have been proven to provide excellent performance for a variety of high-energy physics applications, such as particle identification, event reconstruction and trigger operations. Recently, we developed an end-to-end deep learning approach to identify various particles using low-level detector information from high-energy collisions. These models will be incorporated in the CMS software framework (CMSSW) to enable their use for particle reconstruction or for trigger operation in real time. Incorporating these computational tools in the experimental framework presents new challenges. This paper reports an implementation of the end-to-end deep learning inference with the CMS software framework. The inference has been implemented on GPU for faster computation using ONNX. We have benchmarked the ONNX inference with GPU and CPU using NERSC’s Perlmutter cluster by building a Docker image of the CMS software framework.

Machine learning has been applied to several problems in particle physics research, beginning with applications to high-level physics analysis in the 1990s and 2000s, followed by an explosion of applications in particle and event identification and reconstruction in the 2010s. In this document we discuss promising future research and development areas for machine learning in particle physics. We detail a roadmap for their implementation, software and hardware resource requirements, collaborative initiatives with the data science community, academia and industry, and training the particle physics community in data science. The main objective of the document is to connect and motivate these areas of research and development with the physics drivers of the High-Luminosity Large Hadron Collider and future neutrino experiments and identify the resource needs for their implementation. Additionally we identify areas where collaboration with external communities will be of great benefit.

The first implementation of a Machine Learning Algorithm inside a Level-1 trigger system at the LHC is presented. The Endcap Muon Track Finder (EMTF) at CMS uses Boosted Decision Trees (BDTs) to infer the momentum of muons in the forward region of the detector, based on 25 different variables. Combinations of these variables representing 230 distinct patterns are evaluated offline using regression BDTs. The predictions for the 230 input variable combinations are stored in a 1.2 GB look-up table in the EMTF hardware. The BDTs take advantage of complex correlations between variables, the inhomogeneous magnetic field, and non-linear effects – like inelastic scattering – to distinguish high momentum signal muons from the overwhelming low-momentum background. The new momentum algorithm reduced the background rate by a factor of three with respect to the previous analytic algorithm, with further improvements foreseen in the coming year.

The identity of dark matter remains one of the most pressing questions in physics today. While many promising dark matter candidates have been put forth over the last half-century, to date the true identity of dark matter remains elusive. While it is possible that one of the many proposed candidates may turn out to be dark matter, it is at least equally likely that the correct physical description has yet to be proposed. To address this challenge, novel applications of machine learning can help physicists gain insight into the dark sector from a theory agnostic perspective. In this work we demonstrate the use of unsupervised machine learning techniques to infer the presence of substructure in dark matter halos using galaxy-galaxy strong lensing simulations.

ROOT provides, through TMVA, machine learning tools for data analysis at HEP experiments and beyond. We present recently included features in TMVA and the strategy for future developments in the diversified machine learning landscape. Focus is put on fast machine learning inference, which enables analysts to deploy their machine learning models rapidly on large scale datasets. The new developments are paired with newly designed C++ and Python interfaces supporting modern C++ paradigms and full interoperability in the Python ecosystem. We present as well a new deep learning implementation for convolutional neural network using the cuDNN library for GPU. We show benchmarking results in term of training time and inference time, when comparing with other machine learning libraries such as Keras/Tensorflow.

An essential part of new physics searches at the Large Hadron Collider (LHC) at CERN involves event classification, or distinguishing potential signal events from those coming from background processes. Current machine learning techniques accomplish this using traditional hand-engineered features like particle 4-momenta, motivated by our understanding of particle decay phenomenology. While such techniques have proven useful for simple decays, they are highly dependent on our ability to model all aspects of the phenomenology and detector response. Meanwhile, powerful deep learning algorithms are capable of not only training on high-level features, but of performing feature extraction. In computer vision, convolutional neural networks have become the state-of-the-art for many applications. Motivated by their success, we apply deep learning algorithms to low-level detector data from the 2012 CMS Simulated Open Data to directly learn useful features, in what we call, end-to-end event classification. We demonstrate the power of this approach in the context of a physics search and offer solutions to some of the inherent challenges, such as image construction, image sparsity, combining multiple sub-detectors, and de-correlating the classifier from the search observable, among others.

With the accumulation of large collision datasets at a center-of-mass energy of 13 TeV, the LHC experiments can search for rare processes, where the extraction of signal events from the copious Standard Model backgrounds poses an enormous challenge. Multivariate techniques promise to achieve the best sensitivities by isolating events with higher signal-to-background ratios. Using the search for Higgs bosons decaying to two muons in the CMS experiment as an example, we describe the use of Boosted Decision Trees coupled with automated categorization for optimal event classification, bringing an increase in sensitivity equivalent to 50% more data.

We describe a novel application of the end-to-end deep learning technique to the task of discriminating top quark-initiated jets from those originating from the hadronization of a light quark or a gluon. The end-to-end deep learning technique combines deep learning algorithms and low-level detector representation of the high-energy collision event. In this study, we use lowlevel detector information from the simulated CMS Open Data samples to construct the top jet classifiers. To optimize classifier performance we progressively add low-level information from the CMS tracking detector, including pixel detector reconstructed hits and impact parameters, and demonstrate the value of additional tracking information even when no new spatial structures are added. Relying only on calorimeter energy deposits and reconstructed pixel detector hits, the end-to-end classifier achieves a ROC-AUC score of 0.975±0.002 for the task of classifying boosted top quark jets. After adding derived track quantities, the classifier ROC-AUC score increases to 0.9824±0.0013, serving as the first performance benchmark for these CMS Open Data samples.

Abstract Exoplanets in protoplanetary disks cause localized deviations from Keplerian velocity in channel maps of molecular line emission. Current methods of characterizing these deviations are time consuming,and there is no unified standard approach. We demonstrate that machine learning can quickly and accurately detect the presence of planets. We train our model on synthetic images generated from simulations and apply it to real observations to identify forming planets in real systems. Machine-learning methods, based on computer vision, are not only capable of correctly identifying the presence of one or more planets, but they can also correctly constrain the location of those planets.

We present the activities of the 'New Physics' working group for the 'Physics at TeV Colliders' workshop (Les Houches, France, 1-19 June, 2015). Our report includes new physics studies connected with the Higgs boson and its properties, direct search strategies, reinterpretation of the LHC results in the building of viable models and new computational tool developments. Important signatures for searches for natural new physics at the LHC and new assessments of the interplay between direct dark matter searches and the LHC are also considered.

ROOT is a software framework for large-scale data analysis that provides basic and advanced statistical methods used by high-energy physics experiments. It includes machine learning tools from the ROOT-integrated Toolkit for Multivariate Analysis (TMVA). We present several recent developments in TMVA, including a new modular design, new algorithms for pre-processing, cross-validation, hyperparameter-tuning, deep-learning and interfaces to other machine-learning software packages. TMVA is additionally integrated with Jupyter, making it accessible with a browser.

Accurate and fast simulation of particle physics processes is crucial for the high-energy physics community. Simulating particle interactions with the detector is both time consuming and computationally expensive. With its proton-proton collision energy of 13 TeV, the Large Hadron Collider is uniquely positioned to detect and measure the rare phenomena that can shape our knowledge of new interactions. The High-Luminosity Large Hadron Collider (HLLHC) upgrade will put a significant strain on the computing infrastructure and budget due to increased event rate and levels of pile-up. Simulation of highenergy physics collisions needs to be significantly faster without sacrificing the physics accuracy. Machine learning approaches can offer faster solutions, while maintaining a high level of fidelity. We introduce a graph generative model that provides effiective reconstruction of LHC events on the level of calorimeter deposits and tracks, paving the way for full detector level fast simulation.

Abstract Observations of protoplanetary disks have shown that forming exoplanets leave characteristic imprints on the gas and dust of the disk. In the gas, these forming exoplanets cause deviations from Keplerian motion, which can be detected through molecular line observations. Our previous work has shown that machine learning can correctly determine if a planet is present in these disks. Using our machine-learning models, we identify strong, localized non-Keplerian motion within the disk HD 142666. Subsequent hydrodynamics simulations of a system with a 5 M J planet at 75 au recreate the kinematic structure. By currently established standards in the field, we conclude that HD 142666 hosts a planet. This work represents a first step toward using machine learning to identify previously overlooked non-Keplerian features in protoplanetary disks.

Accurate observation of two or more particles within a very narrow time window has always been a challenge in modern physics. It creates the possibility of correlation experiments, such as the ground-breaking Hanbury Brown-Twiss experiment, leading to new physical insights. For low-energy electrons, one possibility is to use a microchannel plate with subsequent delay lines for the readout of the incident particle hits, a setup called a Delay Line Detector. The spatial and temporal coordinates of more than one particle can be fully reconstructed outside a region called the dead radius. For interesting events, where two electrons are close in space and time, the determination of the individual positions of the electrons requires elaborate peak finding algorithms. While classical methods work well with single particle hits, they fail to identify and reconstruct events caused by multiple nearby particles. To address this challenge, we present a new spatiotemporal machine learning model to identify and reconstruct the position and time of such multi-hit particle signals. This model achieves a much better resolution for nearby particle hits compared to the classical approach, removing some of the artifacts and reducing the dead radius by half. We show that machine learning models can be effective in improving the spatiotemporal performance of delay line detectors.

Abstract The identity of dark matter has remained surprisingly elusive. While terrestrial experiments may be able to nail down a model, an alternative method is to identify dark matter based on astrophysical or cosmological signatures. A particularly sensitive approach is based on the unique signature of dark matter substructure in galaxy–galaxy strong lensing images. Machine-learning applications have been explored for extracting this signal. Because of the limited availability of high-quality strong lensing images, these approaches have exclusively relied on simulations. Due to the differences with the real instrumental data, machine-learning models trained on simulations are expected to lose accuracy when applied to real data. Here domain adaptation can serve as a crucial bridge between simulations and real data applications. In this work, we demonstrate the power of domain adaptation techniques applied to strong gravitational lensing data with dark matter substructure. We show with simulated data sets representative of Euclid and Hubble Space Telescope observations that domain adaptation can significantly mitigate the losses in the model performance when applied to new domains. Lastly, we find similar results utilizing domain adaptation for the problem of lens finding by adapting models trained on a simulated data set to one composed of real lensed and unlensed galaxies from the Hyper Suprime-Cam. This technique can help domain experts build and apply better machine-learning models for extracting useful information from the strong gravitational lensing data expected from the upcoming surveys.

Deep learning techniques have been proven to provide excellent performance for a variety of high-energy physics applications, such as particle identification, event reconstruction and trigger operations. Recently, we developed an end-to-end deep learning approach to identify various particles using low-level detector information from high-energy collisions. These models will be incorporated in the CMS software framework (CMSSW) to enable their use for particle reconstruction or for trigger operation in real-time. Incorporating these computational tools in the experimental framework presents new challenges. This paper reports an implementation of the end-to-end deep learning inference with the CMS software framework. The inference has been implemented on GPU for faster computation using ONNX. We have benchmarked the ONNX inference with GPU and CPU using NERSCs Perlmutter cluster by building a docker image of the CMS software framework.

Machine learning algorithms are heavily relied on to understand the vast amounts of data from high-energy particle collisions at the CERN Large Hadron Collider (LHC). The data from such collision events can naturally be represented with graph structures. Therefore, deep geometric methods, such as graph neural networks (GNNs), have been leveraged for various data analysis tasks in high-energy physics. One typical task is jet tagging, where jets are viewed as point clouds with distinct features and edge connections between their constituent particles. The increasing size and complexity of the LHC particle datasets, as well as the computational models used for their analysis, greatly motivate the development of alternative fast and efficient computational paradigms such as quantum computation. In addition, to enhance the validity and robustness of deep networks, one can leverage the fundamental symmetries present in the data through the use of invariant inputs and equivariant layers. In this paper, we perform a fair and comprehensive comparison between classical graph neural networks (GNNs) and equivariant graph neural networks (EGNNs) and their quantum counterparts: quantum graph neural networks (QGNNs) and equivariant quantum graph neural networks (EQGNN). The four architectures were benchmarked on a binary classification task to classify the parton-level particle initiating the jet. Based on their AUC scores, the quantum networks were shown to outperform the classical networks. However, seeing the computational advantage of the quantum networks in practice may have to wait for the further development of quantum technology and its associated APIs.

This paper presents a comprehensive comparative analysis of the performance of Equivariant Quantum Neural Networks (EQNN) and Quantum Neural Networks (QNN), juxtaposed against their classical counterparts: Equivariant Neural Networks (ENN) and Deep Neural Networks (DNN). We evaluate the performance of each network with two toy examples for a binary classification task, focusing on model complexity (measured by the number of parameters) and the size of the training data set. Our results show that the $\mathbb{Z}_2\times \mathbb{Z}_2$ EQNN and the QNN provide superior performance for smaller parameter sets and modest training data samples.

extended-abstract Share on Accelerating High-energy Physics Exploration with Deep Learning Authors: Dave Ojika University of Florida, Gainesville FL, USA University of Florida, Gainesville FL, USAView Profile , Darin Acosta University of Florida, Gainesville FL, USA University of Florida, Gainesville FL, USAView Profile , Ann Gordon-Ross University of Florida, Gainesville FL, USA University of Florida, Gainesville FL, USAView Profile , Andrew Carnes University of Florida, Gainesville FL, USA University of Florida, Gainesville FL, USAView Profile , Sergei Gleyzer CERN, Geneva, Switzerland CERN, Geneva, SwitzerlandView Profile Authors Info & Claims PEARC17: Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and ImpactJuly 2017 Article No.: 37Pages 1–4https://doi.org/10.1145/3093338.3093340Published:09 July 2017Publication History 0citation126DownloadsMetricsTotal Citations0Total Downloads126Last 12 Months0Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

In high energy physics, variable selection and reduction are key to conducting robust multivariate analyses.Initial variable selection often results in variable sets with greater cardinality than the number of degrees of freedom of the underlying model.This motivates the need for variable reduction, and more fundamentally, for a consistent decision making framework.Such a framework called PARADIGM, based on a global reduction measure called the global loss function and relevant for searches for new phenomena in physics, is described in detail.We illustrate the common pitfalls of variable selection and reduction, such as variable interactions and variable shadowing, and show that PARADIGM gives consistent results in their presence.In this paper, we discuss the application of PARADIGM to several searches for new phenomena in high energy physics and compare the performance of different measures of relative variable importance, in particular of those based on binary regression.Finally, we describe a technique called variable amplification and show how PARADIGM can be used to improve classification performance.

Accurate and fast simulation of particle physics processes is crucial for the high-energy physics community. Simulating particle interactions with detectors is both time consuming and computationally expensive. With the proton-proton collision energy of 13 TeV, the Large Hadron Collider is uniquely positioned to detect and measure the rare phenomena that can shape our knowledge of new interactions. The High-Luminosity Large Hadron Collider (HL-LHC) upgrade will put a significant strain on the computing infrastructure due to increased event rate and levels of pile-up. Simulation of high-energy physics collisions needs to be significantly faster without sacrificing the physics accuracy. Machine learning approaches can offer faster solutions, while maintaining a high level of fidelity. We discuss a graph generative model that provides effective reconstruction of LHC events, paving the way for full detector level fast simulation for HL-LHC.

Machine learning algorithms are gaining ground in high energy physics for applications in particle and event identification, physics analysis, detector reconstruction, simulation and trigger. Currently, most data-analysis tasks at LHC experiments benefit from the use of machine learning. Incorporating these computational tools in the experimental framework presents new challenges. This paper reports on the implementation of the end-to-end deep learning with the CMS software framework and the scaling of the end-to-end deep learning with multiple GPUs. The end-to-end deep learning technique combines deep learning algorithms and low-level detector representation for particle and event identification. We demonstrate the end-to-end implementation on a top quark benchmark and perform studies with various hardware architectures including single and multiple GPUs and Google TPU.

ROOT is a framework for large-scale data analysis that provides basic and advanced statistical methods used by the LHC experiments. These include machine learning algorithms from the ROOT-integrated Toolkit for Multivariate Analysis (TMVA). We present several recent developments in TMVA, including a new modular design, new algorithms for variable importance and cross-validation, interfaces to other machine-learning software packages and integration of TMVA with Jupyter, making it accessible with a browser.

The Toolkit for Multivariate Analysis, TMVA, the machine learning package integrated into the ROOT data analysis framework, has recently seen improvements to its deep learning module, parallelisation of multivariate methods and cross validation. Performance benchmarks on datasets from high-energy physics are presented with a particular focus on the new deep learning module which contains robust fully-connected, convolutional and recurrent deep neural networks implemented on CPU and GPU architectures. Both dense and convo-lutional layers are shown to be competitive on small-scale networks suitable for high-level physics analyses in both training and in single-event evaluation. Par-allelisation efforts show an asymptotical 3-fold reduction in boosted decision tree training time while the cross validation implementation shows significant speed up with parallel fold evaluation.

We describe a novel application of the end-to-end deep learning technique to the task of discriminating top quark-initiated jets from those originating from the hadronization of a light quark or a gluon. The end-to-end deep learning technique uses low-level detector representation of high-energy collision event as inputs to deep learning algorithms. In this study, we use low-level detector information from the simulated Compact Muon Solenoid (CMS) open data samples to construct the top jet classifiers. To optimize classifier performance we progressively add low-level information from the CMS tracking detector, including pixel detector reconstructed hits and impact parameters, and demonstrate the value of additional tracking information even when no new spatial structures are added. Relying only on calorimeter energy deposits and reconstructed pixel detector hits, the end-to-end classifier achieves an area under the receiver operator curve (AUC) score of $0.975\ifmmode\pm\else\textpm\fi{}0.002$ for the task of classifying boosted top quark jets. After adding derived track quantities, the classifier AUC score increases to $0.9824\ifmmode\pm\else\textpm\fi{}0.0013$, serving as the first performance benchmark for these CMS open data samples.

The cross section is one of the most important physical quantities in high-energy physics and the most time consuming to compute. While machine learning has proven to be highly successful in numerical calculations in high-energy physics, analytical calculations using machine learning are still in their infancy. In this work, we use a sequence-to-sequence model, specifically, a transformer, to compute a key element of the cross section calculation, namely, the squared amplitude of an interaction. We show that a transformer model is able to predict correctly 97.6% and 99% of squared amplitudes of QCD and QED processes, respectively, at a speed that is up to orders of magnitude faster than current symbolic computation frameworks. We discuss the performance of the current model, its limitations and possible future directions for this work.

Exoplanets in protoplanetary disks cause localized deviations from Keplerian velocity in channel maps of molecular line emission. Current methods of characterizing these deviations are time consuming, and there is no unified standard approach. We demonstrate that machine learning can quickly and accurately detect the presence of planets. We train our model on synthetic images generated from simulations and apply it to real observations to identify forming planets in real systems. Machine learning methods, based on computer vision, are not only capable of correctly identifying the presence of one or more planets, but they can also correctly constrain the location of those planets.

The first implementation of Boosted Decision Trees (BDTs) inside a Level-1 trigger system at the LHC is presented.The Endcap Muon Track Finder (EMTF) at CMS uses BDTs to infer the momentum of muons in the forward region of the detector, based on 25 different variables.Combinations of these variables are evaluated offline using regression BDTs, whose output is stored in 1.2 GB look-up tables (LUTs) in the EMTF hardware.These BDTs take advantage of complex correlations between variables, the inhomogeneous magnetic field, and non-linear effects such as inelastic scattering to distinguish high-momentum signal muons from the overwhelming low-momentum background.The LUTs are used to turn the complex BDT evaluation into a simple look-up operation in fixed low latency.The new momentum assignment algorithm has reduced the trigger rate by a factor of 3 at the 25 GeV trigger threshold with respect to the legacy system, with further improvements foreseen in the coming year.

For the field of high energy physics to continue to have a bright future, priority within the field must be given to investments in the development of both evolutionary and transformational detector development that is coordinated across the national laboratories and with the university community, international partners and other disciplines. While the fundamental science questions addressed by high energy physics have never been more compelling, there is acute awareness of the challenging budgetary and technical constraints when scaling current technologies. Furthermore, many technologies are reaching their sensitivity limit and new approaches need to be developed to overcome the currently irreducible technological challenges. This situation is unfolding against a backdrop of declining funding for instrumentation, both at the national laboratories and in particular at the universities. This trend has to be reversed for the country to continue to play a leadership role in particle physics, especially in this most promising era of imminent new discoveries that could finally break the hugely successful, but limited, Standard Model of fundamental particle interactions. In this challenging environment it is essential that the community invest anew in instrumentation and optimize the use of the available resources to develop new innovative, cost-effective instrumentation, as this is our best hope to successfully accomplish the mission of high energy physics. This report summarizes the current status of instrumentation for high energy physics, the challenges and needs of future experiments and indicates high priority research areas.

From particle identification to the discovery of the Higgs boson, deep learning algorithms have become an increasingly important tool for data analysis at the Large Hadron Collider (LHC). We present an innovative end-to-end deep learning approach for jet identification at the Compact Muon Solenoid (CMS) experiment at the LHC. The method combines deep neural networks with low-level detector information, such as calorimeter energy deposits and tracking information, to build a discriminator to identify different particle species. Using two physics examples as references: electron vs. photon discrimination and quark vs. gluon discrimination, we demonstrate the performance of the end-to-end approach on simulated events with full detector geometry as available in the CMS Open Data. We also offer insights into the importance of the information extracted from various sub-detectors and describe how end-to-end techniques can be extended to event-level classification using information from the whole CMS detector.

Clues to the identity of dark matter have remained surprisingly elusive, given the scope of experimental programs aimed at its identification. While terrestrial experiments may be able to nail down a model, an alternative, and equally promising, method is to identify dark matter based on astrophysical or cosmological signatures. A particularly sensitive approach is based on the unique signature of dark matter substructure on galaxy-galaxy strong lensing images. Machine learning applications have been explored in detail for extracting just this signal. With limited availability of high quality strong lensing data, these approaches have exclusively relied on simulations. Naturally, due to the differences with the real instrumental data, machine learning models trained on simulations are expected to lose accuracy when applied to real data. This is where domain adaptation can serve as a crucial bridge between simulations and real data applications. In this work, we demonstrate the power of domain adaptation techniques applied to strong gravitational lensing data with dark matter substructure. We show with simulated data sets of varying complexity, that domain adaptation can significantly mitigate the losses in the model performance. This technique can help domain experts build and apply better machine learning models for extracting useful information from strong gravitational lensing data expected from the upcoming surveys.