Matthias Vigl, Nicole Hartman, Michael Kagan
et al.
The success of Large Language Models (LLMs) has established that scaling compute, through joint increases in model capacity and dataset size, is the primary driver of performance in modern machine learning. While machine learning has long been an integral component of High Energy Physics (HEP) data analysis workflows, the compute used to train state-of-the-art HEP models remains orders of magnitude below that of industry foundation models. With scaling laws only beginning to be studied in the field, we investigate neural scaling laws for boosted jet classification using the public JetClass dataset. We derive compute optimal scaling laws and identify an effective performance limit that can be consistently approached through increased compute. We study how data repetition, common in HEP where simulation is expensive, modifies the scaling yielding a quantifiable effective dataset size gain. We then study how the scaling coefficients and asymptotic performance limits vary with the choice of input features and particle multiplicity, demonstrating that increased compute reliably drives performance toward an asymptotic limit, and that more expressive, lower-level features can raise the performance limit and improve results at fixed dataset size.
Machine learning (ML) in high-energy physics (HEP) has moved in the LHC era from an internal detail of experiment software, to an unavoidable public component of many physics data-analyses. Scientific reproducibility thus requires that it be possible to accurately and stably preserve the behaviours of these, sometimes very complex algorithms. We present and document the petrifyML package, which provides missing mechanisms to convert configurations from commonly used HEP ML tools to either the industry-standard ONNX format or to native Python or C++ code, enabling future re-use and re-interpretation of many ML-based experimental studies.
In this contribution we discuss the properties of the jacobian asymmetry, the new observable introduced in hep-ph/2301.04059 for a robust determination of the value and uncertainty of the $W$-boson mass at hadron colliders.
Savannah Thais, Paolo Calafiura, Grigorios Chachamis
et al.
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.
High energy physics (HEP) experiments at the LHC generate data at a rate of $\mathcal{O}(10)$ Terabits per second. This data rate is expected to exponentially increase as experiments will be upgraded in the future to achieve higher collision energies. The increasing size of particle physics datasets combined with the plateauing single-core CPU performance is expected to create a four-fold shortage in computing power by 2030. This makes it necessary to investigate alternate computing architectures to cope with the next generation of HEP experiments. This study provides an overview of different computing techniques used in the LHCb experiment (trigger, track reconstruction, vertex reconstruction, particle identification). Furthermore, this research led to the creation of three event reconstruction algorithms for the LHCb experiment. These algorithms are benchmarked on various computing architectures such as the CPU, GPU, and a new type of processor called the IPU, each roughly containing $\mathcal{O}(10)$, $\mathcal{O}(1000)$, and $\mathcal{O}(1000)$ cores respectively. This research indicates that multi-core architectures such as GPUs and IPUs are better suited for computationally intensive tasks within HEP experiments.