Javier Duarte

Recent results at the Large Hadron Collider (LHC) have pointed to enhanced physics capabilities through the improvement of the real-time event processing techniques. Machine learning methods are ubiquitous and have proven to be very powerful in LHC physics, and particle physics as a whole. However, exploration of the use of such techniques in low-latency, low-power FPGA hardware has only just begun. FPGA-based trigger and data acquisition (DAQ) systems have extremely low, sub-microsecond latency requirements that are unique to particle physics. We present a case study for neural network inference in FPGAs focusing on a classifier for jet substructure which would enable, among many other physics scenarios, searches for new dark sector particles and novel measurements of the Higgs boson. While we focus on a specific example, the lessons are far-reaching. We develop a package based on High-Level Synthesis (HLS) called hls4ml to build machine learning models in FPGAs. The use of HLS increases accessibility across a broad user community and allows for a drastic decrease in firmware development time. We map out FPGA resource usage and latency versus neural network hyperparameters to identify the problems in particle physics that would benefit from performing neural network inference with FPGAs. For our example jet substructure model, we fit well within the available resources of modern FPGAs with a latency on the scale of 100 ns.

A new paradigm for data-driven, model-agnostic new physics searches at colliders is emerging, and aims to leverage recent breakthroughs in anomaly detection and machine learning. In order to develop and benchmark new anomaly detection methods within this framework, it is essential to have standard datasets. To this end, we have created the LHC Olympics 2020, a community challenge accompanied by a set of simulated collider events. Participants in these Olympics have developed their methods using an R&D dataset and then tested them on black boxes: datasets with an unknown anomaly (or not). Methods made use of modern machine learning tools and were based on unsupervised learning (autoencoders, generative adversarial networks, normalizing flows), weakly supervised learning, and semi-supervised learning. This paper will review the LHC Olympics 2020 challenge, including an overview of the competition, a description of methods deployed in the competition, lessons learned from the experience, and implications for data analyses with future datasets as well as future colliders.

We describe the outcome of a data challenge conducted as part of the Dark Machines Initiative and the Les Houches 2019 workshop on Physics at TeV colliders. The challenged aims at detecting signals of new physics at the LHC using unsupervised machine learning algorithms. First, we propose how an anomaly score could be implemented to define model-independent signal regions in LHC searches. We define and describe a large benchmark dataset, consisting of >1 Billion simulated LHC events corresponding to $10~\rm{fb}^{-1}$ of proton-proton collisions at a center-of-mass energy of 13 TeV. We then review a wide range of anomaly detection and density estimation algorithms, developed in the context of the data challenge, and we measure their performance in a set of realistic analysis environments. We draw a number of useful conclusions that will aid the development of unsupervised new physics searches during the third run of the LHC, and provide our benchmark dataset for future studies at https://www.phenoMLdata.org. Code to reproduce the analysis is provided at https://github.com/bostdiek/DarkMachines-UnsupervisedChallenge.

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 We investigate the performance of a jet identification algorithm based on interaction networks (JEDI-net) to identify all-hadronic decays of high-momentum heavy particles produced at the LHC and distinguish them from ordinary jets originating from the hadronization of quarks and gluons. The jet dynamics are described as a set of one-to-one interactions between the jet constituents. Based on a representation learned from these interactions, the jet is associated to one of the considered categories. Unlike other architectures, the JEDI-net models achieve their performance without special handling of the sparse input jet representation, extensive pre-processing, particle ordering, or specific assumptions regarding the underlying detector geometry. The presented models give better results with less model parameters, offering interesting prospects for LHC applications.

We present the implementation of binary and ternary neural networks in the hls4ml library, designed to automatically convert deep neural network models to digital circuits with field-programmable gate arrays (FPGA) firmware. Starting from benchmark models trained with floating point precision, we investigate different strategies to reduce the network's resource consumption by reducing the numerical precision of the network parameters to binary or ternary. We discuss the trade-off between model accuracy and resource consumption. In addition, we show how to balance between latency and accuracy by retaining full precision on a selected subset of network components. As an example, we consider two multiclass classification tasks: handwritten digit recognition with the MNIST data set and jet identification with simulated proton-proton collisions at the CERN Large Hadron Collider. The binary and ternary implementation has similar performance to the higher precision implementation while using drastically fewer FPGA resources.

Abstract We introduce an automated tool for deploying ultra low-latency, low-power deep neural networks with convolutional layers on field-programmable gate arrays (FPGAs). By extending the hls4ml library, we demonstrate an inference latency of 5 µ s using convolutional architectures, targeting microsecond latency applications like those at the CERN Large Hadron Collider. Considering benchmark models trained on the Street View House Numbers Dataset, we demonstrate various methods for model compression in order to fit the computational constraints of a typical FPGA device used in trigger and data acquisition systems of particle detectors. In particular, we discuss pruning and quantization-aware training, and demonstrate how resource utilization can be significantly reduced with little to no loss in model accuracy. We show that the FPGA critical resource consumption can be reduced by 97% with zero loss in model accuracy, and by 99% when tolerating a 6% accuracy degradation.

In general-purpose particle detectors, the particle-flow algorithm may be used to reconstruct a comprehensive particle-level view of the event by combining information from the calorimeters and the trackers, significantly improving the detector resolution for jets and the missing transverse momentum. In view of the planned high-luminosity upgrade of the CERN Large Hadron Collider (LHC), it is necessary to revisit existing reconstruction algorithms and ensure that both the physics and computational performance are sufficient in an environment with many simultaneous proton-proton interactions (pileup). Machine learning may offer a prospect for computationally efficient event reconstruction that is well-suited to heterogeneous computing platforms, while significantly improving the reconstruction quality over rule-based algorithms for granular detectors. We introduce MLPF, a novel, end-to-end trainable, machine-learned particle-flow algorithm based on parallelizable, computationally efficient, and scalable graph neural networks optimized using a multi-task objective on simulated events. We report the physics and computational performance of the MLPF algorithm on a Monte Carlo dataset of top quark-antiquark pairs produced in proton-proton collisions in conditions similar to those expected for the high-luminosity LHC. The MLPF algorithm improves the physics response with respect to a rule-based benchmark algorithm and demonstrates computationally scalable particle-flow reconstruction in a high-pileup environment.

Graph neural networks have been shown to achieve excellent performance for several crucial tasks in particle physics, such as charged particle tracking, jet tagging, and clustering. An important domain for the application of these networks is the FGPA-based first layer of real-time data filtering at the CERN Large Hadron Collider, which has strict latency and resource constraints. We discuss how to design distance-weighted graph networks that can be executed with a latency of less than 1$\mu\mathrm{s}$ on an FPGA. To do so, we consider a representative task associated to particle reconstruction and identification in a next-generation calorimeter operating at a particle collider. We use a graph network architecture developed for such purposes, and apply additional simplifications to match the computing constraints of Level-1 trigger systems, including weight quantization. Using the $\mathtt{hls4ml}$ library, we convert the compressed models into firmware to be implemented on an FPGA. Performance of the synthesized models is presented both in terms of inference accuracy and resource usage.

In this community review report, we discuss applications and techniques for fast machine learning (ML) in science-the concept of integrating powerful ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.

There has been a recent explosion in research into machine-learning-based generative modeling to tackle computational challenges for simulations in high energy physics (HEP). In order to use such alternative simulators in practice, we need well-defined metrics to compare different generative models and evaluate their discrepancy from the true distributions. We present the first systematic review and investigation into evaluation metrics and their sensitivity to failure modes of generative models, using the framework of two-sample goodness-of-fit testing, and their relevance and viability for HEP. Inspired by previous work in both physics and computer vision, we propose two new metrics, the Fr\'echet and kernel physics distances (FPD and KPD, respectively) and perform a variety of experiments measuring their performance on simple Gaussian-distributed and simulated high energy jet datasets. We find FPD, in particular, to be the most sensitive metric to all alternative jet distributions tested and recommend its adoption, along with the KPD and Wasserstein distances between individual feature distributions, for evaluating generative models in HEP. We finally demonstrate the efficacy of these proposed metrics in evaluating and comparing a novel attention-based generative adversarial particle transformer to the state-of-the-art message-passing generative adversarial network jet simulation model. The code for our proposed metrics is provided in the open source jetnet python library.

Abstract A multi-TeV muon collider offers a spectacular opportunity in the direct exploration of the energy frontier. Offering a combination of unprecedented energy collisions in a comparatively clean leptonic environment, a high energy muon collider has the unique potential to provide both precision measurements and the highest energy reach in one machine that cannot be paralleled by any currently available technology. The topic generated a lot of excitement in Snowmass meetings and continues to attract a large number of supporters, including many from the early career community. In light of this very strong interest within the US particle physics community, Snowmass Energy, Theory and Accelerator Frontiers created a cross-frontier Muon Collider Forum in November of 2020. The Forum has been meeting on a monthly basis and organized several topical workshops dedicated to physics, accelerator technology, and detector R&D. Findings of the Forum are summarized in this report.

We develop an algorithm based on an interaction network to identify high-transverse-momentum Higgs bosons decaying to bottom quark-antiquark pairs and distinguish them from ordinary jets that reflect the configurations of quarks and gluons at short distances. The algorithm's inputs are features of the reconstructed charged particles in a jet and the secondary vertices associated with them. Describing the jet shower as a combination of particle-to-particle and particle-to-vertex interactions, the model is trained to learn a jet representation on which the classification problem is optimized. The algorithm is trained on simulated samples of realistic LHC collisions, released by the CMS Collaboration on the CERN Open Data Portal. The interaction network achieves a drastic improvement in the identification performance with respect to state-of-the-art algorithms.

Large-scale particle physics experiments face challenging demands for high-throughput computing resources both now and in the future. New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains. The growing applications of machine learning algorithms in particle physics for simulation, reconstruction, and analysis are naturally deployed on such platforms. We demonstrate that the acceleration of machine learning inference as a web service represents a heterogeneous computing solution for particle physics experiments that potentially requires minimal modification to the current computing model. As examples, we retrain the ResNet-50 convolutional neural network to demonstrate state-of-the-art performance for top quark jet tagging at the LHC and apply a ResNet-50 model with transfer learning for neutrino event classification. Using Project Brainwave by Microsoft to accelerate the ResNet-50 image classification model, we achieve average inference times of 60 (10) ms with our experimental physics software framework using Brainwave as a cloud (edge or on-premises) service, representing an improvement by a factor of approximately 30 (175) in model inference latency over traditional CPU inference in current experimental hardware. A single FPGA service accessed by many CPUs achieves a throughput of 600–700 inferences per second using an image batch of one, comparable to large batch-size GPU throughput and significantly better than small batch-size GPU throughput. Deployed as an edge or cloud service for the particle physics computing model, coprocessor accelerators can have a higher duty cycle and are potentially much more cost-effective.

We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip implementation, hls4ml performs inference of Boosted Decision Tree models with extremely low latency. With a typical latency less than 100 ns, this solution is suitable for FPGA-based real-time processing, such as in the Level-1 Trigger system of a collider experiment. These developments open up prospects for physicists to deploy BDTs in FPGAs for identifying the origin of jets, better reconstructing the energies of muons, and enabling better selection of rare signal processes.

Despite advances in the programmable logic capabilities of modern trigger systems, a significant bottleneck remains in the amount of data to be transported from the detector to off-detector logic where trigger decisions are made. We demonstrate that a neural network autoencoder model can be implemented in a radiation tolerant ASIC to perform lossy data compression alleviating the data transmission problem while preserving critical information of the detector energy profile. For our application, we consider the high-granularity calorimeter from the CMS experiment at the CERN Large Hadron Collider. The advantage of the machine learning approach is in the flexibility and configurability of the algorithm. By changing the neural network weights, a unique data compression algorithm can be deployed for each sensor in different detector regions, and changing detector or collider conditions. To meet area, performance, and power constraints, we perform a quantization-aware training to create an optimized neural network hardware implementation. The design is achieved through the use of high-level synthesis tools and the hls4ml framework, and was processed through synthesis and physical layout flows based on a LP CMOS 65 nm technology node. The flow anticipates 200 Mrad of ionizing radiation to select gates, and reports a total area of 3.6 mm^2 and consumes 95 mW of power. The simulated energy consumption per inference is 2.4 nJ. This is the first radiation tolerant on-detector ASIC implementation of a neural network that has been designed for particle physics applications.

To study the physics of fundamental particles and their interactions, the Large Hadron Collider was constructed at CERN, where protons collide to create new particles measured by detectors. Collisions occur at a frequency of 40 MHz, and with an event size of roughly 1 MB it is impossible to read out and store the generated amount of data from the detector and therefore a multi-tiered, real-time filtering system is required. In this paper, we show how to adapt and deploy deep-learning-based autoencoders for the unsupervised detection of new physics signatures in the challenging environment of a real-time event selection system at the Large Hadron Collider. The first-stage filter, implemented on custom electronics, decides within a few microseconds whether an event should be kept or discarded. At this stage, the rate is reduced from 40 MHz to about 100 kHz. We demonstrate the deployment of an unsupervised selection algorithm on this custom electronics, running in as little as 80 ns and enhancing the signal-over-background ratio by three orders of magnitude. This work enables the practical deployment of these networks during the next data-taking campaign of the Large Hadron Collider. The Large Hadron Collider produces 40 million collision events per second, most of which need to be discarded by a real-time filtering system. Unsupervised deep learning algorithms are developed and deployed on custom electronics to search for rare events indicating new physics, rather than for specific events led by theory.

The determination of charged particle trajectories in collisions at the CERN Large Hadron Collider (LHC) is an important but challenging problem, especially in the high interaction density conditions expected during the future high-luminosity phase of the LHC (HL-LHC). Graph neural networks (GNNs) are a type of geometric deep learning algorithm that has successfully been applied to this task by embedding tracker data as a graph-nodes represent hits, while edges represent possible track segments-and classifying the edges as true or fake track segments. However, their study in hardware- or software-based trigger applications has been limited due to their large computational cost. In this paper, we introduce an automated translation workflow, integrated into a broader tool called hls4ml, for converting GNNs into firmware for field-programmable gate arrays (FPGAs). We use this translation tool to implement GNNs for charged particle tracking, trained using the TrackML challenge dataset, on FPGAs with designs targeting different graph sizes, task complexites, and latency/throughput requirements. This work could enable the inclusion of charged particle tracking GNNs at the trigger level for HL-LHC experiments.

To enable the reusability of massive scientific datasets by humans and machines, researchers aim to adhere to the principles of findability, accessibility, interoperability, and reusability (FAIR) for data and artificial intelligence (AI) models. This article provides a domain-agnostic, step-by-step assessment guide to evaluate whether or not a given dataset meets these principles. We demonstrate how to use this guide to evaluate the FAIRness of an open simulated dataset produced by the CMS Collaboration at the CERN Large Hadron Collider. This dataset consists of Higgs boson decays and quark and gluon background, and is available through the CERN Open Data Portal. We use additional available tools to assess the FAIRness of this dataset, and incorporate feedback from members of the FAIR community to validate our results. This article is accompanied by a Jupyter notebook to visualize and explore this dataset. This study marks the first in a planned series of articles that will guide scientists in the creation of FAIR AI models and datasets in high energy particle physics.

There has been significant work recently in developing machine learning models in high energy physics (HEP), for tasks such as classification, simulation, and anomaly detection. Typically, these models are adapted from those designed for datasets in computer vision or natural language processing without necessarily incorporating inductive biases suited to HEP data, such as respecting its inherent symmetries. Such inductive biases can make the model more performant and interpretable, and reduce the amount of training data needed. To that end, we develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant with respect to the proper, orthochronous Lorentz group $\mathrm{SO}^+(3,1)$, with a latent space living in the representations of the group. We present our architecture and several experimental results on jets at the LHC and find it significantly outperforms a non-Lorentz-equivariant graph neural network baseline on compression and reconstruction, and anomaly detection. We also demonstrate the advantage of such an equivariant model in analyzing the latent space of the autoencoder, which can have a significant impact on the explainability of anomalies found by such black-box machine learning models.

There has been significant work recently in developing machine learning (ML) models in high energy physics (HEP) for tasks such as classification, simulation, and anomaly detection. Often these models are adapted from those designed for datasets in computer vision or natural language processing, which lack inductive biases suited to HEP data, such as equivariance to its inherent symmetries. Such biases have been shown to make models more performant and interpretable, and reduce the amount of training data needed. To that end, we develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant with respect to the proper, orthochronous Lorentz group $\mathrm{SO}^+(3,1)$, with a latent space living in the representations of the group. We present our architecture and several experimental results on jets at the LHC and find it outperforms graph and convolutional neural network baseline models on several compression, reconstruction, and anomaly detection metrics. We also demonstrate the advantage of such an equivariant model in analyzing the latent space of the autoencoder, which can improve the explainability of potential anomalies discovered by such ML models.

A foundational set of findable, accessible, interoperable, and reusable (FAIR) principles were proposed in 2016 as prerequisites for proper data management and stewardship, with the goal of enabling the reusability of scholarly data. The principles were also meant to apply to other digital assets, at a high level, and over time, the FAIR guiding principles have been re-interpreted or extended to include the software, tools, algorithms, and workflows that produce data. FAIR principles are now being adapted in the context of AI models and datasets. Here, we present the perspectives, vision, and experiences of researchers from different countries, disciplines, and backgrounds who are leading the definition and adoption of FAIR principles in their communities of practice, and discuss outcomes that may result from pursuing and incentivizing FAIR AI research. The material for this report builds on the FAIR for AI Workshop held at Argonne National Laboratory on June 7, 2022.

Deep neural networks use skip connections to improve training convergence. However, these skip connections are costly in hardware, requiring extra buffers and increasing on- and off-chip memory utilization and bandwidth requirements. In this article, we show that skip connections can be optimized for hardware when tackled with a hardware-software codesign approach. We argue that while a network’s skip connections are needed for the network to learn, they can later be removed or shortened to provide a more hardware-efficient implementation with minimal to no accuracy loss. We introduce Tailor , a codesign tool whose hardware-aware training algorithm gradually removes or shortens a fully trained network’s skip connections to lower the hardware cost. Tailor improves resource utilization by up to 34% for block random access memories (BRAMs), 13% for flip-flops (FFs), and 16% for look-up tables (LUTs) for on-chip, dataflow-style architectures. Tailor increases performance by 30% and reduces memory bandwidth by 45% for a two-dimensional processing element array architecture.

Updates over the previous version: updated validation outputs for the cluster-based model fixed a bug with how the PF candidates were stored added single particle gun samples to validation added new timing runs for the baseline algo, included memory information run the GNN model up to ~10k inputs added hypertuning summary tables Trained models and evaluation results for the upcoming paper "Improved particle-flow event reconstruction with scalable neural networks for current and future particle detectors", https://doi.org/10.48550/arXiv.2309.06782. The archive contains the following subfolders: clusters_best_tuned_gnn_clic_v130 MLPF GNN model configs and weight files after hypertuning the inputs are reconstructed tracks and Pandora clusters the outputs are reconstructed PF candidates trained on tt and qq v1.3.0 (1M events each) hits MLPF GNN model configs and weight files inputs are reconstructed tracks and calorimeter hits outputs are reconstructed PF candidates trained on tt, qq and gun samples (K0L, gamma, pi+-, pi0, neutron, ele, mu) v1.2.0 training was restarted several times from previous checkpoints hypertuning GNN and transformer model before and after hypertuning summary tables of the hypertuning runs timing scaling study of baseline PF with number of gun particles on CPU scaling study of GNN model with number of input elements on GPU gpu_scaling the scaling study of model training on multiple accelerator cards The training dataset is available at Pata, Joosep, Wulff, Eric, Duarte, Javier, Mokhtar, Farouk, Zhang, Mengke, Girone, Maria, & Southwick, David. (2023). Simulated datasets for detector and particle flow reconstruction: CLIC detector (1.1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.8260741

We investigate how to improve new physics detection strategies exploiting variational autoencoders and normalizing flows for anomaly detection at the Large Hadron Collider. As a working example, we consider the DarkMachines challenge dataset. We show how different design choices (e.g., event representations, anomaly score definitions, network architectures) affect the result on specific benchmark new physics models. Once a baseline is established, we discuss how to improve the anomaly detection accuracy by exploiting normalizing flow layers in the latent space of the variational autoencoder.

Machine learning methods have a long history of applications in high energy physics (HEP). Recently, there is a growing interest in exploiting these methods to reconstruct particle signatures from raw detector data. In order to benefit from modern deep learning algorithms that were initially designed for computer vision or natural language processing tasks, it is common practice to transform HEP data into images or sequences. Conversely, graph neural networks (GNNs), which operate on graph data composed of elements with a set of features and their pairwise connections, provide an alternative way of incorporating weight sharing, local connectivity, and specialized domain knowledge. Particle physics data, such as the hits in a tracking detector, can generally be represented as graphs, making the use of GNNs natural. In this chapter, we recapitulate the mathematical formalism of GNNs and highlight aspects to consider when designing these networks for HEP data, including graph construction, model architectures, learning objectives, and graph pooling. We also review promising applications of GNNs for particle tracking and reconstruction in HEP and summarize the outlook for their deployment in current and future experiments.

There has been a recent explosion in research into machine-learning-based generative modeling to tackle computational challenges for simulations in high energy physics (HEP). In order to use such alternative simulators in practice, we need well defined metrics to compare different generative models and evaluate their discrepancy from the true distributions. We present the first systematic review and investigation into evaluation metrics and their sensitivity to failure modes of generative models, using the framework of two-sample goodness-of-fit testing, and their relevance and viability for HEP. Inspired by previous work in both physics and computer vision, we propose two new metrics, the Fr\'echet and kernel physics distances (FPD and KPD), and perform a variety of experiments measuring their performance on simple Gaussian-distributed, and simulated high energy jet datasets. We find FPD, in particular, to be the most sensitive metric to all alternative jet distributions tested and recommend its adoption, along with the KPD and Wasserstein distances between individual feature distributions, for evaluating generative models in HEP. We finally demonstrate the efficacy of these proposed metrics in evaluating and comparing a novel attention-based generative adversarial particle transformer to the state-of-the-art message-passing generative adversarial network jet simulation model.

Machine learning methods have a long history of applications in high energy physics (HEP). Recently, there is a growing interest in exploiting these methods to reconstruct particle signatures from raw detector data. In order to benefit from modern deep learning algorithms that were initially designed for computer vision or natural language processing tasks, it is common practice to transform HEP data into images or sequences. Conversely, graph neural networks (GNNs), which operate on graph data composed of elements with a set of features and their pairwise connections, provide an alternative way of incorporating weight sharing, local connectivity, and specialized domain knowledge. Particle physics data, such as the hits in a tracking detector, can generally be represented as graphs, making the use of GNNs natural. In this chapter, we recapitulate the mathematical formalism of GNNs and highlight aspects to consider when designing these networks for HEP data, including graph construction, model architectures, learning objectives, and graph pooling. We also review promising applications of GNNs for particle tracking and reconstruction in HEP and summarize the outlook for their deployment in current and future experiments.

We provide details on the implementation of a machine-learning based particle flow algorithm for CMS. The standard particle flow algorithm reconstructs stable particles based on calorimeter clusters and tracks to provide a global event reconstruction that exploits the combined information of multiple detector subsystems, leading to strong improvements for quantities such as jets and missing transverse energy. We have studied a possible evolution of particle flow towards heterogeneous computing platforms such as GPUs using a graph neural network. The machine-learned PF model reconstructs particle candidates based on the full list of tracks and calorimeter clusters in the event. For validation, we determine the physics performance directly in the CMS software framework when the proposed algorithm is interfaced with the offline reconstruction of jets and missing transverse energy. We also report the computational performance of the algorithm, which scales approximately linearly in runtime and memory usage with the input size.

We describe a method for precisely regulating the gradient magnet power supply (GMPS) at the Fermilab Booster accelerator complex using a neural network trained via reinforcement learning. We demonstrate preliminary results by training a surrogate machine-learning model on real accelerator data to emulate the GMPS, and using this surrogate model in turn to train the neural network for its regulation task. We additionally show how the neural networks to be deployed for control purposes may be compiled to execute on field-programmable gate arrays (FPGAs), and show the first machine-learning based control algorithm implemented on an FPGA for controls at the Fermilab accelerator complex. As there are no surprise latencies on an FPGA, this capability is important for operational stability in complicated environments such as an accelerator facility.

Recent work has demonstrated that geometric deep learning methods such as graph neural networks (GNNs) are well suited to address a variety of reconstruction problems in high energy particle physics. In particular, particle tracking data is naturally represented as a graph by identifying silicon tracker hits as nodes and particle trajectories as edges; given a set of hypothesized edges, edge-classifying GNNs identify those corresponding to real particle trajectories. In this work, we adapt the physics-motivated interaction network (IN) GNN toward the problem of particle tracking in pileup conditions similar to those expected at the high-luminosity Large Hadron Collider. Assuming idealized hit filtering at various particle momenta thresholds, we demonstrate the IN's excellent edge-classification accuracy and tracking efficiency through a suite of measurements at each stage of GNN-based tracking: graph construction, edge classification, and track building. The proposed IN architecture is substantially smaller than previously studied GNN tracking architectures; this is particularly promising as a reduction in size is critical for enabling GNN-based tracking in constrained computing environments. Furthermore, the IN may be represented as either a set of explicit matrix operations or a message passing GNN. Efforts are underway to accelerate each representation via heterogeneous computing resources towards both high-level and low-latency triggering applications.

In the next decade, the demands for computing in large scientific experiments are expected to grow tremendously. During the same time period, CPU performance increases will be limited. At the CERN Large Hadron Collider (LHC), these two issues will confront one another as the collider is upgraded for high luminosity running. Alternative processors such as graphics processing units (GPUs) can resolve this confrontation provided that algorithms can be sufficiently accelerated. In many cases, algorithmic speedups are found to be largest through the adoption of deep learning algorithms. We present a comprehensive exploration of the use of GPU-based hardware acceleration for deep learning inference within the data reconstruction workflow of high energy physics. We present several realistic examples and discuss a strategy for the seamless integration of coprocessors so that the LHC can maintain, if not exceed, its current performance throughout its running.

Efficient machine learning implementations optimized for inference in hardware have wide-ranging benefits, depending on the application, from lower inference latency to higher data throughput and reduced energy consumption. Two popular techniques for reducing computation in neural networks are pruning, removing insignificant synapses, and quantization, reducing the precision of the calculations. In this work, we explore the interplay between pruning and quantization during the training of neural networks for ultra low latency applications targeting high energy physics use cases. Techniques developed for this study have potential applications across many other domains. We study various configurations of pruning during quantization-aware training, which we term quantization-aware pruning , and the effect of techniques like regularization, batch normalization, and different pruning schemes on performance, computational complexity, and information content metrics. We find that quantization-aware pruning yields more computationally efficient models than either pruning or quantization alone for our task. Further, quantization-aware pruning typically performs similar to or better in terms of computational efficiency compared to other neural architecture search techniques like Bayesian optimization. Surprisingly, while networks with different training configurations can have similar performance for the benchmark application, the information content in the network can vary significantly, affecting its generalizability.

We study how to use deep variational autoencoders (VAEs) for a fast simulation of jets of particles at the Large Hadron Collider. We represent jets as a list of constituents, characterized by their momenta. Starting from a simulation of the jet before detector effects, we train a deep VAE to return the corresponding list of constituents after detection. Doing so, we bypass both the time-consuming detector simulation and the collision reconstruction steps of a traditional processing chain, speeding up significantly the events generation workflow. Through model optimization and hyperparameter tuning, we achieve state-of-the-art precision on the jet four-momentum, while providing an accurate description of the constituents momenta, and an inference time comparable to that of a rule-based fast simulation.

A new paradigm for data-driven, model-agnostic new physics searches at colliders is emerging, and aims to leverage recent breakthroughs in anomaly detection and machine learning. In order to develop and benchmark new anomaly detection methods within this framework, it is essential to have standard datasets. To this end, we have created the LHC Olympics 2020, a community challenge accompanied by a set of simulated collider events. Participants in these Olympics have developed their methods using an R&D dataset and then tested them on black boxes: datasets with an unknown anomaly (or not). This paper will review the LHC Olympics 2020 challenge, including an overview of the competition, a description of methods deployed in the competition, lessons learned from the experience, and implications for data analyses with future datasets as well as future colliders.

We present test beam studies and results on the timing performance and characterization of the time resolution of Lutetium–Yttrium Orthosilicate (LYSO)-based calorimeters. We demonstrate that a time resolution of 30 ps is achievable for a particular design. Furthermore, we discuss precision timing calorimetry as a tool for the mitigation of physics object performance degradation effects due to the large number of simultaneous interactions in the high luminosity environment foreseen at the Large Hadron Collider.

We develop and study FPGA implementations of algorithms for charged particle tracking based on graph neural networks. The two complementary FPGA designs are based on OpenCL, a framework for writing programs that execute across heterogeneous platforms, and hls4ml, a high-level-synthesis-based compiler for neural network to firmware conversion. We evaluate and compare the resource usage, latency, and tracking performance of our implementations based on a benchmark dataset. We find a considerable speedup over CPU-based execution is possible, potentially enabling such algorithms to be used effectively in future computing workflows and the FPGA-based Level-1 trigger at the CERN Large Hadron Collider.

In high energy physics (HEP), jets are collections of correlated particles produced ubiquitously in particle collisions such as those at the CERN Large Hadron Collider (LHC). Machine learning (ML)-based generative models, such as generative adversarial networks (GANs), have the potential to significantly accelerate LHC jet simulations. However, despite jets having a natural representation as a set of particles in momentum-space, a.k.a. a particle cloud, there exist no generative models applied to such a dataset. In this work, we introduce a new particle cloud dataset (JetNet), and apply to it existing point cloud GANs. Results are evaluated using (1) 1-Wasserstein distances between high- and low-level feature distributions, (2) a newly developed Fr\'{e}chet ParticleNet Distance, and (3) the coverage and (4) minimum matching distance metrics. Existing GANs are found to be inadequate for physics applications, hence we develop a new message passing GAN (MPGAN), which outperforms existing point cloud GANs on virtually every metric and shows promise for use in HEP. We propose JetNet as a novel point-cloud-style dataset for the ML community to experiment with, and set MPGAN as a benchmark to improve upon for future generative models. Additionally, to facilitate research and improve accessibility and reproducibility in this area, we release the open-source JetNet Python package with interfaces for particle cloud datasets, implementations for evaluation and loss metrics, and more tools for ML in HEP development.

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. These architectures provide an initial design for models that could be used for tagging at the CERN LHC during its high-luminosity phase. The high-luminosity upgrade will lead to a five-fold increase in its instantaneous luminosity for proton-proton collisions and, in turn, higher data volume and complexity, such as the availability of jet constituents. Through quantization-aware training and efficient hardware implementations, we show that O(100) ns inference of complex architectures such as deep sets and interaction networks is feasible at a low computational resource cost.

This study introduces a novel transformer model optimized for large-scale point cloud processing in scientific domains such as high-energy physics (HEP) and astrophysics. Addressing the limitations of graph neural networks and standard transformers, our model integrates local inductive bias and achieves near-linear complexity with hardware-friendly regular operations. One contribution of this work is the quantitative analysis of the error-complexity tradeoff of various sparsification techniques for building efficient transformers. Our findings highlight the superiority of using locality-sensitive hashing (LSH), especially OR \& AND-construction LSH, in kernel approximation for large-scale point cloud data with local inductive bias. Based on this finding, we propose LSH-based Efficient Point Transformer (\textbf{HEPT}), which combines E$^2$LSH with OR \& AND constructions and is built upon regular computations. HEPT demonstrates remarkable performance in two critical yet time-consuming HEP tasks, significantly outperforming existing GNNs and transformers in accuracy and computational speed, marking a significant advancement in geometric deep learning and large-scale scientific data processing. Our code is available at \url{https://github.com/Graph-COM/HEPT}.

With more scientific fields relying on neural networks (NNs) to process data incoming at extreme throughputs and latencies, it is crucial to develop NNs with all their parameters stored on-chip. In many of these applications, there is not enough time to go off-chip and retrieve weights. Even more so, off-chip memory such as DRAM does not have the bandwidth required to process these NNs as fast as the data is being produced (e.g., every 25 ns). As such, these extreme latency and bandwidth requirements have architectural implications for the hardware intended to run these NNs: 1) all NN parameters must fit on-chip, and 2) codesigning custom/reconfigurable logic is often required to meet these latency and bandwidth constraints. In our work, we show that many scientific NN applications must run fully on chip, in the extreme case requiring a custom chip to meet such stringent constraints.

Abstract Efficient and accurate algorithms are necessary to reconstruct particles in the highly granular detectors anticipated at the High-Luminosity Large Hadron Collider and the Future Circular Collider. We study scalable machine learning models for event reconstruction in electron-positron collisions based on a full detector simulation. Particle-flow reconstruction can be formulated as a supervised learning task using tracks and calorimeter clusters. We compare a graph neural network and kernel-based transformer and demonstrate that we can avoid quadratic operations while achieving realistic reconstruction. We show that hyperparameter tuning significantly improves the performance of the models. The best graph neural network model shows improvement in the jet transverse momentum resolution by up to 50% compared to the rule-based algorithm. The resulting model is portable across Nvidia, AMD and Habana hardware. Accurate and fast machine-learning based reconstruction can significantly improve future measurements at colliders.

A rastreabilidade dos produtos, bem como o uso de selos de certificação, tornaram-se essenciais para satisfazer as expectativas dos consumidores por segurança, sustentabilidade, qualidade e transparência nas cadeias de suprimentos. Atualmente, a maioria dos sistemas de informação dessas cadeias opera com sistemas centralizados, ou explora a tecnologia blockchain como uma solução potencial para aprimorar a integridade e rastreabilidade na cadeia. No entanto, desafios como fragmentação de dados, vulnerabilidades, custos de implementação e preocupações com privacidade de dados são obstáculos para essas soluções. Dessa forma, o artigo explora o uso das Identidades Autossoberanas (SSI), especificamente no projeto COTTONTRUST, que adota a SSI para melhorar a integridade e rastreabilidade da cadeia algodoeira. O foco está na análise dos tempos de transação para criar identidades autossoberanas da cadeia do algodão, avaliando a escalabilidade do COTTONTRUST.

The findable, accessible, interoperable, and reusable (FAIR) data principles serve as a framework for examining, evaluating, and improving data sharing to advance scientific endeavors. There is an emerging trend to adapt these principles for machine learning models—algorithms that learn from data without specific coding—and, more generally, AI models, due to AI’s swiftly growing impact on scientific and engineering sectors. In this paper, we propose a practical definition of the FAIR principles for AI models and provide a template program for their adoption. We exemplify this strategy with an implementation from high-energy physics, where a graph neural network is employed to detect Higgs bosons decaying into two bottom quarks.

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.

Abstract At the High Luminosity Large Hadron Collider (HL-LHC), traditional track reconstruction techniques that are critical for physics analysis will need to be upgraded to scale with track density. Quantum annealing has shown promise in its ability to solve combinatorial optimization problems amidst an ongoing effort to establish evidence of a quantum speedup. As a step towards exploiting such potential speedup, we investigate a track reconstruction approach by adapting the existing geometric Denby-Peterson (Hopfield) network method to the quantum annealing framework for HL-LHC conditions. We develop additional techniques to embed the problem onto existing and near-term quantum annealing hardware. Results using simulated annealing and quantum annealing with the D-Wave 2X system on the TrackML open dataset are presented, demonstrating the successful application of a quantum annealing algorithm to the track reconstruction challenge. We find that combinatorial optimization problems can effectively reconstruct tracks, suggesting possible applications for fast hardware-specific implementations at the HL-LHC while leaving open the possibility of a quantum speedup for tracking.

Many physical systems can be best understood as sets of discrete data with associated relationships. Where previously these sets of data have been formulated as series or image data to match the available machine learning architectures, with the advent of graph neural networks (GNNs), these systems can be learned natively as graphs. This allows a wide variety of high- and low-level physical features to be attached to measurements and, by the same token, a wide variety of HEP tasks to be accomplished by the same GNN architectures. GNNs have found powerful use-cases in reconstruction, tagging, generation and end-to-end analysis. With the wide-spread adoption of GNNs in industry, the HEP community is well-placed to benefit from rapid improvements in GNN latency and memory usage. However, industry use-cases are not perfectly aligned with HEP and much work needs to be done to best match unique GNN capabilities to unique HEP obstacles. We present here a range of these capabilities, predictions of which are currently being well-adopted in HEP communities, and which are still immature. We hope to capture the landscape of graph techniques in machine learning as well as point out the most significant gaps that are inhibiting potentially large leaps in research.

Computing needs for high energy physics are already intensive and are expected to increase drastically in the coming years. In this context, heterogeneous computing, specifically as-a-service computing, has the potential for significant gains over traditional computing models. Although previous studies and packages in the field of heterogeneous computing have focused on GPUs as accelerators, FPGAs are an extremely promising option as well. A series of workflows are developed to establish the performance capabilities of FPGAs as a service. Multiple different devices and a range of algorithms for use in high energy physics are studied. For a small, dense network, the throughput can be improved by an order of magnitude with respect to GPUs as a service. For large convolutional networks, the throughput is found to be comparable to GPUs as a service. This work represents the first open-source FPGAs-as-a-service toolkit.

We describe the outcome of a data challenge conducted as part of the Dark Machines Initiative and the Les Houches 2019 workshop on Physics at TeV colliders. The challenged aims at detecting signals of new physics at the LHC using unsupervised machine learning algorithms. First, we propose how an anomaly score could be implemented to define model-independent signal regions in LHC searches. We define and describe a large benchmark dataset, consisting of >1 Billion simulated LHC events corresponding to $10~\rm{fb}^{-1}$ of proton-proton collisions at a center-of-mass energy of 13 TeV. We then review a wide range of anomaly detection and density estimation algorithms, developed in the context of the data challenge, and we measure their performance in a set of realistic analysis environments. We draw a number of useful conclusions that will aid the development of unsupervised new physics searches during the third run of the LHC, and provide our benchmark dataset for future studies at this https URL. Code to reproduce the analysis is provided at this https URL.

Current and future high energy physics particle colliders are capable to provide instantaneous luminosities of 1034 cm-2s-1 and above. The high center of mass energy, the large number of simultaneous collision of beam particles in the experiments and the very high repetition rates of the collision events pose huge challenges. They result in extremely high particle fluxes, causing very high occupancies in the particle physics detectors operating at these machines. To reconstruct the physics events, the detectors have to make as much information as possible available on the final state particles. We discuss how timing information with a precision of around 10 ps and below can aid the reconstruction of the physics events under such challenging conditions. High energy photons play a crucial role in this context. About one third of the particle flux originating from high energy hadron collisions is detected as photons, stemming from the decays of neutral mesons. In addition, many key physics signatures under study are identified by high energy photons in the final state. They pose a particular challenge in that they can only be detected once they convert in the detector material. The particular challenge in measuring the time of arrival of a high energy photon lies in the stochastic component of the distance to the initial conversion and the size of the electromagnetic shower. They extend spatially over distances which propagation times of the initial photon and the subsequent electromagnetic shower which are large compared to the desired precision. We present studies and measurements from test beams and a cosmic muon test stand for calorimeter based timing measurements to explore the ultimate timing precision achievable for high energy photons of 10 GeV and above. We put particular focus on techniques to measure the timing with a precision of about 10 ps in association with the energy of the photon. For calorimeters utilizing scintillating materials and light guiding components, the propagation speed of the scintillation light in the calorimeter is important. We present studies and measurements of the propagation speed on a range of detector geometries. Finally, possible applications of precision timing in future high energy physics experiments are discussed.

There are different models which are based on the gauge symmetry <svg style="vertical-align:-3.24037pt;width:180.66251px;" id="M2" height="14.725" version="1.1" viewBox="0 0 180.66251 14.725" width="180.66251" xmlns="http://www.w3.org/2000/svg"> <g transform="matrix(1.25,0,0,-1.25,0,14.725)"> <g transform="translate(72,-60.22)"> <text transform="matrix(1,0,0,-1,-71.95,63.5)"> <tspan style="font-size: 12.50px; " x="0" y="0">𝑆</tspan> <tspan style="font-size: 12.50px; " x="8.1269503" y="0">𝑈</tspan> <tspan style="font-size: 12.50px; " x="18.304392" y="0">(</tspan> <tspan style="font-size: 12.50px; " x="22.467892" y="0">3</tspan> <tspan style="font-size: 12.50px; " x="28.719391" y="0">)</tspan> </text> <text transform="matrix(1,0,0,-1,-39.05,60.37)"> <tspan style="font-size: 8.75px; " x="0" y="0">𝐶</tspan> </text> <text transform="matrix(1,0,0,-1,-29.21,63.5)"> <tspan style="font-size: 12.50px; " x="0" y="0">⊗</tspan> <tspan style="font-size: 12.50px; " x="13.303192" y="0">𝑆</tspan> <tspan style="font-size: 12.50px; " x="21.430141" y="0">𝑈</tspan> <tspan style="font-size: 12.50px; " x="31.620087" y="0">(</tspan> <tspan style="font-size: 12.50px; " x="35.783585" y="0">3</tspan> <tspan style="font-size: 12.50px; " x="42.035088" y="0">)</tspan> </text> <text transform="matrix(1,0,0,-1,17,60.37)"> <tspan style="font-size: 8.75px; " x="0" y="0">𝐿</tspan> </text> <text transform="matrix(1,0,0,-1,26.47,63.5)"> <tspan style="font-size: 12.50px; " x="0" y="0">⊗</tspan> <tspan style="font-size: 12.50px; " x="13.303192" y="0">𝑈</tspan> <tspan style="font-size: 12.50px; " x="23.493137" y="0">(</tspan> <tspan style="font-size: 12.50px; " x="27.656635" y="0">1</tspan> <tspan style="font-size: 12.50px; " x="33.908134" y="0">)</tspan> </text> <text transform="matrix(1,0,0,-1,64.54,60.37)"> <tspan style="font-size: 8.75px; " x="0" y="0">𝑋</tspan> </text> </g> </g> </svg> (331), and some of them include exotic particles, and others are constructed without any exotic charges assigned to the fermionic spectrum. Each model build-up on 331 symmetry has its own interesting properties according to the representations of the gauge group used for the fermionic spectrum, that is, the main reason to explore and identify the possible sources of flavor changing neutral currents and lepton flavor violation at tree level.

At the High Luminosity Large Hadron Collider (HL-LHC), traditional track reconstruction techniques that are critical for analysis are expected to face challenges due to scaling with track density. Quantum annealing has shown promise in its ability to solve combinatorial optimization problems amidst an ongoing effort to establish evidence of a quantum speedup. As a step towards exploiting such potential speedup, we investigate a track reconstruction approach by adapting the existing geometric Denby-Peterson (Hopfield) network method to the quantum annealing framework and to HL-LHC conditions. Furthermore, we develop additional techniques to embed the problem onto existing and near-term quantum annealing hardware. Results using simulated annealing and quantum annealing with the D-Wave 2X system on the TrackML dataset are presented, demonstrating the successful application of a quantum annealing-inspired algorithm to the track reconstruction challenge. We find that combinatorial optimization problems can effectively reconstruct tracks, suggesting possible applications for fast hardware-specific implementations at the LHC while leaving open the possibility of a quantum speedup for tracking.

Objective This study aimed to determine the prevalence of dizziness and its associated factors in patients with COM at two otologic referral centres in a middle-income country.Design Cross-sectional study. Adults with and without COM diagnosis from two otology-referral centres in Bogotá (Colombia) were included. Dizziness and quality of life were assessed using the “Chronic Suppurative Otitis Media Questionnaire-12” (COMQ-12), and sociodemographic questionnaires were applied. Otoscopic evaluation and audiometric data were collected.Study sample A total of 231 adults.Results Of the 231 participants, up to 64.5% (n = 149) reported at least mild inconvenience due to dizziness. Factors associated with dizziness included female sex (aPR: 1.23; 95% CI: 1.04–1.46), chronic suppurative otitis media (aPR: 3.02; 95% CI: 1.21–7.52), and severe tinnitus (aPR: 1.75; 95% CI: 1.24–2.48). An interaction was found between socioeconomic status and educational level, with more frequent reports of dizziness in the middle/high economic status and secondary education (aPR: 3.09; 95% CI: 0.52–18.55; p < 0.001). Differences of 14 points in symptom severity and 18.5 points in the total score of the COMQ-12 were found between the groups with dizziness and without dizziness.Conclusions Dizziness was frequent in patients with COM and was associated with severe tinnitus and quality of life deterioration.

The particle-flow (PF) algorithm, which infers particles based on tracks and calorimeter clusters, is of central importance to event reconstruction in the CMS experiment at the CERN LHC, and has been a focus of development in light of planned Phase-2 running conditions with an increased pileup and detector granularity. In recent years, the machine learned particle-flow (MLPF) algorithm, a graph neural network that performs PF reconstruction, has been explored in CMS, with the possible advantages of directly optimizing for the physical quantities of interest, being highly reconfigurable to new conditions, and being a natural fit for deployment to heterogeneous accelerators. We discuss progress in CMS towards an improved implementation of the MLPF reconstruction, now optimized using generator/simulation-level particle information as the target for the first time. This paves the way to potentially improving the detector response in terms of physical quantities of interest. We describe the simulation-based training target, progress and studies on event-based loss terms, details on the model hyperparameter tuning, as well as physics validation with respect to the current PF algorithm in terms of high-level physical quantities such as the jet and missing transverse momentum resolutions. We find that the MLPF algorithm, trained on a generator/simulator level particle information for the first time, results in broadly compatible particle and jet reconstruction performance with the baseline PF, setting the stage for improving the physics performance by additional training statistics and model tuning.

Voyager is an innovative computational resource designed by the San Diego Supercomputer Center in collaboration with technology partners to accelerate the development and performance of artificial intelligence and machine learning applications in science and engineering. Based on Intel's Habana Labs first-generation deep learning (Gaudi) training and (Goya) inference processors, Voyager is funded by the National Science Foundation's Advanced Computing Systems & Services Program as a Category II system and will be operated for 5 years, starting with an initial 3-year exploratory test-bed phase that will be followed by a 2-year allocated production phase for the national research community. Its AI-focused hardware features several innovative components, including fully-programmable tensor processing cores, high-bandwidth memory, and integrated, on-chip RDMA over Converged Ethernet network interfaces. In addition, Habana's SynapseAI software suite provides seamless integration to popular machine learning frameworks like PyTorch and TensorFlow for end users. Here, we describe the design motivation for Voyager, its system architecture, software and user environment, initial benchmarking results, and the early science use cases and applications currently being ported to and deployed on the system.

Abstract The findable, accessible, interoperable, and reusable (FAIR) data principles provide a framework for examining, evaluating, and improving how data is shared to facilitate scientific discovery. Generalizing these principles to research software and other digital products is an active area of research. Machine learning models—algorithms that have been trained on data without being explicitly programmed—and more generally, artificial intelligence (AI) models, are an important target for this because of the ever-increasing pace with which AI is transforming scientific domains, such as experimental high energy physics (HEP). In this paper, we propose a practical definition of FAIR principles for AI models in HEP and describe a template for the application of these principles. We demonstrate the template’s use with an example AI model applied to HEP, in which a graph neural network is used to identify Higgs bosons decaying to two bottom quarks. We report on the robustness of this FAIR AI model, its portability across hardware architectures and software frameworks, and its interpretability.

<strong>Data description</strong> Datasets generated using Key4HEP and the CLIC detector model suitable for particle flow reconstruction studies. The datasets contain generator particles, reconstructed tracks and calorimeter hits, reconstructed Pandora PF particles and their respective links in the EDM4HEP format. The following processes have been simulated with Pythia 8: p8_ee_tt_ecm380: ee -&gt; ttbar, center of mass energy at 380 GeV p8_ee_qq_ecm380: ee -&gt; Z* -&gt; qqbar, center of mass energy at 380 GeV p8_ee_ZH_Htautau: ee -&gt; ZH -&gt; Higgs decaying to tau leptons, center of mass energy at 380 GeV p8_ee_WW_fullhad: ee -&gt; WW -&gt; W decaying hadronically, center of mass energy at 380 GeV p8_ee_tt_ecm380_PU10: ee -&gt; ttbar with on average 10 Poisson-distributed events from ee-&gt;gg overlayed, center of mass energy at 380 GeV The following single particle gun samples have been generated with ddsim: e+/e-: single electron with energy between 1 and 100 GeV mu+/mu-: single muon with energy between 1 and 100 GeV kaon0L: single K0L with energy between 1 and 100 GeV neutron: single neutron with energy between 1 and 100 GeV pi+/pi-: single charged pion with energy between 1 and 100 GeV pi0: single neutral pion with energy between 1 and 100 GeV gamma: single photon with energy between 1 and 100 GeV The detector simulation has been done with Geant4, the reconstruction with Marlin interfaced via Key4HEP which includes PF reconstruction with Pandora, all using publicly available models and code. <strong>Contents</strong> This record includes the following files: *_10files.tar: small archives of 10 files for each data sample, suitable for testing dataset_full.txt: the full list of files, hosted at the Julich HPC courtesy of the Raise CoE project, ~2.5TB total *.cmd: the Pythia8 cards pythia.py: the pythia steering code for Key4HEP run_sim.sh: the steering script for generating, simulating and reconstructing a single file of 100 events from the p8_ee_tt_ecm380, p8_ee_qq_ecm380, p8_ee_ZH_Htautau, p8_ee_WW_fullhad datasets run_sim_pu.sh: the steering script for generating, simulating and reconstructing a single file of 100 events from the p8_ee_tt_ecm380_PU10 dataset run_sim_gun.sh: the steering script for generating the single-particle gun samples run_sim_gun_np.sh: the steering script for generating multi-particle gun samples (extensive datasets have not yet been generated) check_files.py: the main driver script that configures the full statistics and creates submission scripts for all the simulations PandoraSettings.zip: the settings used for Pandora PF reconstruction main19.cc: the Pythia8+HepMC driver code for generating the events with PU overlay clicRec_e4h_input.py: the steering configuration of the reconstruction modules in Key4HEP clic_steer.py: the steering configuration of the Geant4 simulation modules in Key4HEP clic-visualize.ipynb: an example notebook demonstrating how the dataset can be loaded and events visualized in Python visualization.mp4: an example visualization of the hits and generator particles of a single ttbar event from the dataset <strong>Dataset semantics</strong> Each file consists of event records. Each event contains structured branches of the relevant physics data. The branches relevant to particle flow reconstruction include: MCParticles: the ground truth generator particles ECALBarrel, ECALEndcap, ECALOther, HCALBarrel, HCALEndcap, HCALOther, MUON: reconstructed hits in the various calorimeter subsystems SiTracks_Refitted: the reconstructed tracks PandoraClusters: the calorimeter hits, clustered by Pandora to calorimeter clusters MergedRecoParticles: the reconstructed particles from the Pandora particle flow algorithm CalohitMCTruthLink: the links between MC particles and reconstructed calorimeter hits SiTracksMCTruthLink: the links between MC particles and reconstructed tracks The full details of the EDM4HEP format are available here. <strong>Dataset characteristics</strong> The full dataset in dataset_full.txt consists of 43 tar files of up to 100GB each. The tar files contain in total 58068 files, 2.5TB in the ROOT EDM4HEP format. The subset in *_10files.tar for consists of 150 files, 26GB in the ROOT EDM4HEP format. <strong>How can you use these data?</strong> The ROOT files can be directly loaded with the uproot Python library. <strong>Disclaimer</strong> These are simulated samples suitable for conceptual machine learning R&amp;D and software performance studies. They have not been calibrated with respect to real data, and should not be used to derive physics projections about the detectors. Neither CLIC nor CERN endorse any works, scientific or otherwise, produced using these data. All releases will have a unique DOI that you are requested to cite in any applications or publications.

We develop a graph generative adversarial network to generate sparse data sets like those produced at the CERN Large Hadron Collider (LHC). We demonstrate this approach by training on and generating sparse representations of MNIST handwritten digit images and jets of particles in proton-proton collisions like those at the LHC. We find the model successfully generates sparse MNIST digits and particle jet data. We quantify agreement between real and generated data with a graph-based Fréchet Inception distance, and the particle and jet feature-level 1-Wasserstein distance for the MNIST and jet datasets respectively.

Particle colliders operating at high luminosities present challenging environments for high energy physics event reconstruction and analysis. We discuss how timing information, with a precision on the order of 10 ps, can aid in the reconstruction of physics events under such conditions. We present calorimeter based timing measurements from test beam experiments in which we explore the ultimate timing precision achievable for high energy photons or electrons of 10 GeV and above. Using a prototype calorimeter consisting of a 1.7×1.7×1.7 cm3 lutetium–yttrium oxyortho-silicate (LYSO) crystal cube, read out by micro-channel plate photomultipliers, we demonstrate a time resolution of 33.5±2.1 ps for an incoming beam energy of 32 GeV. In a second measurement, using a 2.5×2.5×20 cm3 LYSO crystal placed perpendicularly to the electron beam, we achieve a time resolution of 59±11 ps using a beam energy of 4 GeV. We also present timing measurements made using a shashlik-style calorimeter cell made of LYSO and tungsten plates, and demonstrate that the apparatus achieves a time resolution of 54±5 ps for an incoming beam energy of 32 GeV.

Dataset of high-pT jets from simulations of LHC proton-proton collisions Prepared for FastML/HLS4ML studies: https://fastmachinelearning.org Includes: High level features (see https://arxiv.org/abs/1804.06913) Images: jet images with up to 150 particles/jet (see https://arxiv.org/abs/1908.05318) List: list of jet features with up to 150 particles/jet (see https://arxiv.org/abs/1908.05318)

We present extensions to the Open Neural Network Exchange (ONNX) intermediate representation format to represent arbitrary-precision quantized neural networks. We first introduce support for low precision quantization in existing ONNX-based quantization formats by leveraging integer clipping, resulting in two new backward-compatible variants: the quantized operator format with clipping and quantize-clip-dequantize (QCDQ) format. We then introduce a novel higher-level ONNX format called quantized ONNX (QONNX) that introduces three new operators -- Quant, BipolarQuant, and Trunc -- in order to represent uniform quantization. By keeping the QONNX IR high-level and flexible, we enable targeting a wider variety of platforms. We also present utilities for working with QONNX, as well as examples of its usage in the FINN and hls4ml toolchains. Finally, we introduce the QONNX model zoo to share low-precision quantized neural networks.

We present our development experience and recent results for the MLPerf Tiny Inference Benchmark on field-programmable gate array (FPGA) platforms. We use the open-source hls4ml and FINN workflows, which aim to democratize AI-hardware codesign of optimized neural networks on FPGAs. We present the design and implementation process for the keyword spotting, anomaly detection, and image classification benchmark tasks. The resulting hardware implementations are quantized, configurable, spatial dataflow architectures tailored for speed and efficiency and introduce new generic optimizations and common workflows developed as a part of this work. The full workflow is presented from quantization-aware training to FPGA implementation. The solutions are deployed on system-on-chip (Pynq-Z2) and pure FPGA (Arty A7-100T) platforms. The resulting submissions achieve latencies as low as 20 $\mu$s and energy consumption as low as 30 $\mu$J per inference. We demonstrate how emerging ML benchmarks on heterogeneous hardware platforms can catalyze collaboration and the development of new techniques and more accessible tools.

Steganography and digital watermarking are the tasks of hiding recoverable data in image pixels. Deep neural network (DNN) based image steganography and watermarking techniques are quickly replacing traditional hand-engineered pipelines. DNN based watermarking techniques have drastically improved the message capacity, imperceptibility and robustness of the embedded watermarks. However, this improvement comes at the cost of increased computational overhead of the watermark encoder neural network. In this work, we design the first accelerator platform FastStamp to perform DNN based steganography and digital watermarking of images on hardware. We first propose a parameter efficient DNN model for embedding recoverable bit-strings in image pixels. Our proposed model can match the success metrics of prior state-of-the-art DNN based watermarking methods while being significantly faster and lighter in terms of memory footprint. We then design an FPGA based accelerator framework to further improve the model throughput and power consumption by leveraging data parallelism and customized computation paths. FastStamp allows embedding hardware signatures into images to establish media authenticity and ownership of digital media. Our best design achieves 68 times faster inference as compared to GPU implementations of prior DNN based watermark encoder while consuming less power.

Gluon (<code>g</code>), Top Quark (<code>t</code>), Light Quark (<code>q</code>), W boson (<code>w</code>), and Z boson (<code>z</code>) jets of ~1 TeV transverse momentum (\(p_T\)), as introduced in Ref. [1]. Each file has <code>particle_features</code>; and <code>jet_features</code>; arrays, containing the list of particles' features per jet and the corresponding jet's features, respectively. <code>particle_features</code> is of shape [N, 30, 4], where N is the total number of jets, 30 is the max number of particles per jet, and 4 is the number of particle features, in order: [\(\eta_{\mathrm{rel}}\), \(\varphi_{\mathrm{rel}}\), \(p_{T\mathrm{rel}}\), mask]. See [1, Sec. 2] for definitions of these. <code>jet_features</code> is of shape [N, 4], where 4 is the number of jet features, in order: [\(p_T\), \(\eta\), mass, # of particles]. N = 177252, 177945, 170679, 177172, 176952 for g, t, q, w, z jets respectively. Finally, the JetNet package is recommended for easy access and use of this dataset. [1] Kansal et. al., <em>Particle Cloud Generation with Message Passing Generative Adversarial Networks</em>, NeurIPS 2021 <code>arXiv:2106.11535</code>

Scintillator based calorimeter technology is studied with the aim to achieve particle detection with a time resolution on the order of a few 10 ps for photons and electrons at energies of a few GeV and above. We present results from a prototype of a 1.4×1.4×11.4 cm3 sampling calorimeter cell consisting of tungsten absorber plates and Cerium-doped Lutetium Yttrium Orthosilicate (LYSO) crystal scintillator plates. The LYSO plates are read out with wave lengths shifting fibers which are optically coupled to fast photo detectors on both ends of the fibers. The measurements with electrons were performed at the Fermilab Test Beam Facility (FTBF) and the CERN SPS H2 test beam. In addition to the baseline setup plastic scintillation counter and a MCP-PMT were used as trigger and as a reference for a time of flight measurement (TOF). We also present measurements with a fast laser to further characterize the response of the prototype and the photo sensors. All data were recorded using a DRS4 fast sampling digitizer. These measurements are part of an R&D program whose aim is to demonstrate the feasibility of building a large scale electromagnetic calorimeter with a time resolution on the order of 10 ps, to be used in high energy physics experiments.

We discuss a model-independent strategy for boosting new physics searches with the help of an unsupervised anomaly detection algorithm. Prior to a search, each input event is preprocessed by the algorithm - a variational autoencoder (VAE). Based on the loss assigned to each event, input data can be split into a background control sample and a signal enriched sample. Following this strategy, one can enhance the sensitivity to new physics with no assumption on the underlying new physics signature. Our results show that a typical BSM search on the signal enriched group is more sensitive than an equivalent search on the original dataset.

This article presents a comparative study using two global optimization algorithms, Electromagnetic Field Optimization (EFO) and Heat Transfer Search (HTS). These techniques are efficient alternatives when classical methods find limitations to solve real problems. To verify methods performance, the rectangular microchannel heat sink design was implemented formulating the respective&nbsp; Inverse Heat Transfer Problem (IHTP). Experimental results were competitively compared with the traditional Levenberg-Marquardt (LM) outcomes. Moreover, global algorithms achieved estimations with errors lower than 5%, and they converged at least three times faster than LM. &nbsp

Machine learning methods are ubiquitous and have proven to be very powerful in LHC physics, and particle physics as a whole. However, exploration of such techniques in low-latency, low-power FPGA (Field Programmable Gate Array) hardware has only just begun. FPGA-based trigger and data acquisition systems have extremely low, sub-microsecond latency requirements that are unique to particle physics. We present a case study for neural network inference in FPGAs focusing on a classifier for jet substructure which would enable many new physics measurements. While we focus on a specific example, the lessons are far-reaching. A compiler package is developed based on High-Level Synthesis (HLS) called HLS4ML to build machine learning models in FPGAs. The use of HLS increases accessibility across a broad user community and allows for a drastic decrease in firmware development time. We map out FPGA resource usage and latency versus neural network hyperparameters to allow for directed resource tuning in the low latency environment and assess the impact on our benchmark Physics performance scenario For our example jet substructure model, we fit well within the available resources of modern FPGAs with latency on the scale of 100~ns.

Common and community software packages, such as ROOT, Geant4 and event generators have been a key part of the LHC's success so far and continued development and optimisation will be critical in the future. The challenges are driven by an ambitious physics programme, notably the LHC accelerator upgrade to high-luminosity, HL-LHC, and the corresponding detector upgrades of ATLAS and CMS. In this document we address the issues for software that is used in multiple experiments (usually even more widely than ATLAS and CMS) and maintained by teams of developers who are either not linked to a particular experiment or who contribute to common software within the context of their experiment activity. We also give space to general considerations for future software and projects that tackle upcoming challenges, no matter who writes it, which is an area where community convergence on best practice is extremely useful.

Dataset of high-pT jets from simulations of LHC proton-proton collisions Prepared for FastML/HLS4ML studies: https://fastmachinelearning.org Includes: High level features (see https://arxiv.org/abs/1804.06913) Images: jet images with up to 30 particles/jet (see https://arxiv.org/abs/1908.05318) List: list of jet features with up to 30 particles/jet (see https://arxiv.org/abs/1908.05318)

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.Moreover, exploiting symmetries inherent in physics data have inspired physics-informed ML as a vibrant sub-field of computer science research.HEP researchers benefit greatly from materials widely available materials for use in education, training and workforce development.They are also contributing to these materials and providing software to DS/ML-related fields.Increasingly, physics departments are offering courses at the intersection of DS, ML and physics, often using curricula developed by HEP researchers and involving open software and data used in HEP.In this white paper, we explore synergies between HEP research and DS/ML education, discuss opportunities and challenges at this intersection, and propose community activities that will be mutually beneficial.

Data scouting, introduced by CMS in 2011, is the use of specialized data streams based on reduced event content, enabling LHC experiments to record unprecedented numbers of proton-proton collision events that would otherwise be rejected by the usual filters. These streams were created to maintain sensitivity to new light resonances decaying to jets or muons, while requiring minimal online and offline resources, and taking advantage of the fast and accurate online reconstruction algorithms of the high-level trigger. The viability of this technique was demonstrated by CMS in 2012, when 18.8 fb$^{-1}$ of collision data at $\sqrt{s} = 8$ TeV were collected and analyzed. For LHC Run 2, CMS, ATLAS, and LHCb implemented or expanded similar reduced-content data streams, promoting the concept to an essential and flexible discovery tool for the LHC.

Accessible machine learning algorithms, software, and diagnostic tools for energy-efficient devices and systems are extremely valuable across a broad range of application domains. In scientific domains, real-time near-sensor processing can drastically improve experimental design and accelerate scientific discoveries. To support domain scientists, we have developed hls4ml, an open-source software-hardware codesign workflow to interpret and translate machine learning algorithms for implementation with both FPGA and ASIC technologies. We expand on previous hls4ml work by extending capabilities and techniques towards low-power implementations and increased usability: new Python APIs, quantization-aware pruning, end-to-end FPGA workflows, long pipeline kernels for low power, and new device backends include an ASIC workflow. Taken together, these and continued efforts in hls4ml will arm a new generation of domain scientists with accessible, efficient, and powerful tools for machine-learning-accelerated discovery.

Autoencoders have useful applications in high energy physics in anomaly detection, particularly for jets - collimated showers of particles produced in collisions such as those at the CERN Large Hadron Collider. We explore the use of graph-based autoencoders, which operate on jets in their "particle cloud" representations and can leverage the interdependencies among the particles within a jet, for such tasks. Additionally, we develop a differentiable approximation to the energy mover's distance via a graph neural network, which may subsequently be used as a reconstruction loss function for autoencoders.

We develop a generative neural network for the generation of sparse data in particle physics using a permutation-invariant and physics-informed loss function. The input dataset used in this study consists of the particle constituents of hadronic jets due to its sparsity and the possibility of evaluating the network's ability to accurately describe the particles and jets properties. A variational autoencoder composed of convolutional layers in the encoder and decoder is used as the generator. The loss function consists of a reconstruction error term and the Kullback-Leibler divergence between the output of the encoder and the latent vector variables. The permutation-invariant loss on the particles' properties is combined with two mean-squared error terms that measure the difference between input and output jets mass and transverse momentum, which improves the network's generation capability as it imposes physics constraints, allowing the model to learn the kinematics of the jets.

There are different models based on the gauge symmetry $SU(3)_C \otimes SU(3)_L \otimes U(1)_X$ (331) some of them includes exotic particles and others are constructed without any exotic charges assigned to the fermionic spectrum. Each model build up on 331 symmetry has its own interesting properties according to the representations of the gauge group used for the fermionic spectrum; that is the main reason to explore and identify the possible sources of flavor changing neutral currents and lepton flavor violation at tree level.

Computing plays a significant role in all areas of high energy physics. The Snowmass 2021 CompF4 topical group's scope is facilities R&D, where we consider "facilities" as the computing hardware and software infrastructure inside the data centers plus the networking between data centers, irrespective of who owns them, and what policies are applied for using them. In other words, it includes commercial clouds, federally funded High Performance Computing (HPC) systems for all of science, and systems funded explicitly for a given experimental or theoretical program. This topical group report summarizes the findings and recommendations for the storage, processing, networking and associated software service infrastructures for future high energy physics research, based on the discussions organized through the Snowmass 2021 community study.

Deep neural networks use skip connections to improve training convergence. However, these skip connections are costly in hardware, requiring extra buffers and increasing on- and off-chip memory utilization and bandwidth requirements. In this paper, we show that skip connections can be optimized for hardware when tackled with a hardware-software codesign approach. We argue that while a network's skip connections are needed for the network to learn, they can later be removed or shortened to provide a more hardware efficient implementation with minimal to no accuracy loss. We introduce Tailor, a codesign tool whose hardware-aware training algorithm gradually removes or shortens a fully trained network's skip connections to lower their hardware cost. Tailor improves resource utilization by up to 34% for BRAMs, 13% for FFs, and 16% for LUTs for on-chip, dataflow-style architectures. Tailor increases performance by 30% and reduces memory bandwidth by 45% for a 2D processing element array architecture.

Deep neural networks employ skip connections – identity functions that combine the outputs of different layers-to improve training convergence; however, these skip connections are costly to implement in hardware. In particular, for inference accelerators on resource-limited platforms, they require extra buffers, increasing not only on- and off-chip memory utilization but also memory bandwidth requirements. Thus, a network that has skip connections costs more to deploy in hardware than one that has none. We argue that, for certain classification tasks, a network's skip connections are needed for the network to learn but not necessary for inference after convergence. We thus explore removing skip connections from a fully-trained network to mitigate their hardware cost. From this investigation, we introduce a fine-tuning/retraining method that adapts a network's skip connections – by either removing or shortening them-to make them fit better in hardware with minimal to no loss in accuracy. With these changes, we decrease resource utilization by up to 34% for BRAMs, 7% for FFs, and 12% LUTs when implemented on an FPGA.

We develop an end-to-end workflow for the training and implementation of co-designed neural networks (NNs) for efficient field-programmable gate array (FPGA) and application-specific integrated circuit (ASIC) hardware. Our approach leverages Hessian-aware quantization (HAWQ) of NNs, the Quantized Open Neural Network Exchange (QONNX) intermediate representation, and the hls4ml tool flow for transpiling NNs into FPGA and ASIC firmware. This makes efficient NN implementations in hardware accessible to nonexperts, in a single open-sourced workflow that can be deployed for real-time machine learning applications in a wide range of scientific and industrial settings. We demonstrate the workflow in a particle physics application involving trigger decisions that must operate at the 40 MHz collision rate of the CERN Large Hadron Collider (LHC). Given the high collision rate, all data processing must be implemented on custom ASIC and FPGA hardware within a strict area and latency. Based on these constraints, we implement an optimized mixed-precision NN classifier for high-momentum particle jets in simulated LHC proton-proton collisions.

The Earth mover's distance (EMD) is a useful metric for image recognition and classification, but its usual implementations are not differentiable or too slow to be used as a loss function for training other algorithms via gradient descent. In this paper, we train a convolutional neural network (CNN) to learn a differentiable, fast approximation of the EMD and demonstrate that it can be used as a substitute for computing-intensive EMD implementations. We apply this differentiable approximation in the training of an autoencoder-inspired neural network (encoder NN) for data compression at the high-luminosity LHC at CERN. The goal of this encoder NN is to compress the data while preserving the information related to the distribution of energy deposits in particle detectors. We demonstrate that the performance of our encoder NN trained using the differentiable EMD CNN surpasses that of training with loss functions based on mean squared error.

Modern large-scale physics experiments create datasets with sizes and streaming rates that can exceed those from industry leaders such as Google Cloud and Netflix. Fully processing these datasets requires both sufficient compute power and efficient workflows. Recent advances in Machine Learning (ML) and Artificial Intelligence (AI) can either improve or replace existing domain-specific algorithms to increase workflow efficiency. Not only can these algorithms improve the physics performance of current algorithms, but they can often be executed more quickly, especially when run on coprocessors such as GPUs or FPGAs. In the winter of 2023, MIT hosted the Accelerating Physics with ML at MIT workshop, which brought together researchers from gravitational-wave physics, multi-messenger astrophysics, and particle physics to discuss and share current efforts to integrate ML tools into their workflows. The following white paper highlights examples of algorithms and computing frameworks discussed during this workshop and summarizes the expected computing needs for the immediate future of the involved fields.

In-time particle trajectory reconstruction in the Large Hadron Collider is challenging due to the high collision rate and numerous particle hits. Using GNN (Graph Neural Network) on FPGA has enabled superior accuracy with flexible trajectory classification. However, existing GNN architectures have inefficient resource usage and insufficient parallelism for edge classification. This paper introduces a resource-efficient GNN architecture on FPGAs for low latency particle tracking. The modular architecture facilitates design scalability to support large graphs. Leveraging the geometric properties of hit detectors further reduces graph complexity and resource usage. Our results on Xilinx UltraScale+ VU9P demonstrate 1625x and 1574x performance improvement over CPU and GPU respectively.

With the increase in demand for food, agricultural activity began to be characterized by standardized monoculture systems and began to be carried out in an intensified manner, however, due to the high demand for natural resources, it has been presenting itself with low environmental sustainability.

Efficient and accurate algorithms are necessary to reconstruct particles in the highly granular detectors anticipated at the High-Luminosity Large Hadron Collider and the Future Circular Collider. We study scalable machine learning models for event reconstruction in electron-positron collisions based on a full detector simulation. Particle-flow reconstruction can be formulated as a supervised learning task using tracks and calorimeter clusters. We compare a graph neural network and kernel-based transformer and demonstrate that we can avoid quadratic operations while achieving realistic reconstruction. We show that hyperparameter tuning significantly improves the performance of the models. The best graph neural network model shows improvement in the jet transverse momentum resolution by up to 50% compared to the rule-based algorithm. The resulting model is portable across Nvidia, AMD and Habana hardware. Accurate and fast machine-learning based reconstruction can significantly improve future measurements at colliders.

[This corrects the article DOI: 10.3389/fdata.2022.787421.].

Abstract In high energy physics, one of the most important processes for collider data analysis is the comparison of collected and simulated data. Nowadays the state-of-the-art for data generation is in the form of Monte Carlo (MC) generators. However, because of the upcoming high-luminosity upgrade of the Large Hadron Collider (LHC), there will not be enough computational power or time to match the amount of needed simulated data using MC methods. An alternative approach under study is the usage of machine learning generative methods to fulfill that task. Since the most common final-state objects of high-energy proton collisions are hadronic jets, which are collections of particles collimated in a given region of space, this work aims to develop a convolutional variational autoencoder (ConVAE) for the generation of particle-based LHC hadronic jets. Given the ConVAE’s limitations, a normalizing flow (NF) network is coupled to it in a two-step training process, which shows improvements on the results for the generated jets. The ConVAE+NF network is capable of generating a jet in <?CDATA $18.30 \pm 0.04\,\,{\mu\text{s}}$?> <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mn>18.30</mml:mn> <mml:mo>±</mml:mo> <mml:mn>0.04</mml:mn> <mml:mrow> <mml:mi>μ</mml:mi> <mml:mtext>s</mml:mtext> </mml:mrow> </mml:math> , making it one of the fastest methods for this task up to now.

In high energy physics, one of the most important processes for collider data analysis is the comparison of collected and simulated data. Nowadays the state-of-the-art for data generation is in the form of Monte Carlo (MC) generators. However, because of the upcoming high-luminosity upgrade of the LHC, there will not be enough computational power or time to match the amount of needed simulated data using MC methods. An alternative approach under study is the usage of machine learning generative methods to fulfill that task.Since the most common final-state objects of high-energy proton collisions are hadronic jets, which are collections of particles collimated in a given region of space, this work aims to develop a convolutional variational autoencoder (ConVAE) for the generation of particle-based LHC hadronic jets. Given the ConVAE's limitations, a normalizing flow (NF) network is coupled to it in a two-step training process, which shows improvements on the results for the generated jets. The ConVAE+NF network is capable of generating a jet in $18.30 \pm 0.04 \ \mu$s, making it one of the fastest methods for this task up to now.

JetNet is a Python package that aims to increase accessibility and reproducibility for machine learning (ML) research in high energy physics (HEP), primarily related to particle jets.Based on the popular PyTorch ML framework, it provides easy-to-access and standardized interfaces for multiple heterogeneous HEP datasets and implementations of evaluation metrics, loss functions, and more general utilities relevant to HEP.

In-time particle trajectory reconstruction in the Large Hadron Collider is challenging due to the high collision rate and numerous particle hits. Using GNN (Graph Neural Network) on FPGA has enabled superior accuracy with flexible trajectory classification. However, existing GNN architectures have inefficient resource usage and insufficient parallelism for edge classification. This paper introduces a resource-efficient GNN architecture on FPGAs for low latency particle tracking. The modular architecture facilitates design scalability to support large graphs. Leveraging the geometric properties of hit detectors further reduces graph complexity and resource usage. Our results on Xilinx UltraScale+ VU9P demonstrate 1625x and 1574x performance improvement over CPU and GPU respectively.

Uma identidade digital única no ciberespaço é essencial para identificar dispositivos, pessoas ou objetos. O modelo SSI é um novo paradigma descentralizado que permite às entidades gerenciar suas próprias identidades de forma confiável, eliminando a dependência de organizações certificadoras. Nesse contexto, este trabalho explora o uso do modelo SSI no ecossistema que envolve a cadeia produtiva do algodão, modelando a entidade UBA. Com isso, busca analisar como as identidades autossoberanas podem ser efetivamente utilizadas para promover a confiança e a transparência na cadeia, com potencial para resolver desafios relacionados à autenticidade, certificação e rastreabilidade do algodão.

Abstract We study scalable machine learning models for full event reconstruction in high-energy electron-positron collisions based on a highly granular detector simulation. Particle-flow (PF) reconstruction can be formulated as a supervised learning task using tracks and calorimeter clusters or hits. We compare a graph neural network and kernel-based transformer and demonstrate that both avoid quadratic memory allocation and computational cost while achieving realistic PF reconstruction. We show that hyperparameter tuning on a supercomputer significantly enhances the physics performance of the models, improving the jet transverse momentum resolution by up to 50% compared to the baseline. The resulting model is highly portable across hardware processors, supporting Nvidia, AMD, and Intel Habana cards. Finally, we demonstrate that the model can be trained on highly granular inputs consisting of tracks and calorimeter hits, resulting in a competitive physics performance with the baseline. Datasets and software to reproduce the studies are published following the findable, accessible, interoperable, and reusable (FAIR) principles.

We present FKeras, an open-source tool that uses Hessian information to quickly find which parameters in a neural network are sensitive to radiation faults, reducing the usual 200% resource overhead needed to protect them.

Abstract The Earth mover’s distance (EMD) is a useful metric for image recognition and classification, but its usual implementations are not differentiable or too slow to be used as a loss function for training other algorithms via gradient descent. In this paper, we train a convolutional neural network (CNN) to learn a differentiable, fast approximation of the EMD and demonstrate that it can be used as a substitute for computing-intensive EMD implementations. We apply this differentiable approximation in the training of an autoencoder-inspired neural network (encoder NN) for data compression at the high-luminosity LHC at CERN The goal of this encoder NN is to compress the data while preserving the information related to the distribution of energy deposits in particle detectors. We demonstrate that the performance of our encoder NN trained using the differentiable EMD CNN surpasses that of training with loss functions based on mean squared error.