Search NASA⌕ Search

SEARCH · Search NASA

Results for “Generative Machine Learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 883 records · Page 49
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

An Advanced Machine Learning and Artificial Intelligence System for Demonstrating Radiation Regulatory Compliance in DOE Accelerator Facilities

In this Phase II proposal, Applied Research LLC (ARLLC), Thomas Jefferson National Accelerator Facility (Jefferson Lab), and Old Dominion University (ODU) propose the combination of domain knowledge (beam characteristics, fixed structural shielding, earthen burden (the soil and foliage added to the dome of the experimental halls as additional shielding), etc.), machine learning (ML) and/or artificial intelligence (AI) to correlate a variety of multi-modal onsite signals and the radiation fields seen in accessible areas of the accelerator site and the site boundary. The ML/AI will consider the complex influence of environmental parameters affecting the radon contribution of the measurements, focusing on actual data obtained from Jefferson Lab. In Phase I, the coded beam and location data were fed into a deep learning model to predict doses at several designated locations in Jefferson Lab’s facility. Moreover, a dense radiation map was generated using only a sparse collection of the samples in a facility. In Phase II, we will develop a software prototype containing a radiation prediction algorithm, dense radiation map algorithms, and background noise prediction algorithms, with actual data used to evaluate the prototype. This work will provide a framework for evaluation of radiation measurement results around the site based on learned responses. In addition, the proposed approach allows more granular mapping of radiation levels. Better understanding and communication of these levels is related to the overall approach in keeping doses to personnel ALARA.

43 PARTICLE ACCELERATORS↗
NASA NTRS2005

Acquisition of Autonomous Behaviors by Robotic Assistants

Our research achievements under the NASA-JSC grant contributed significantly in the following areas. Multi-agent based robot control architecture called the Intelligent Machine Architecture (IMA) : The Vanderbilt team received a Space Act Award for this research from NASA JSC in October 2004. Cognitive Control and the Self Agent : Cognitive control in human is the ability to consciously manipulate thoughts and behaviors using attention to deal with conflicting goals and demands. We have been updating the IMA Self Agent towards this goal. If opportunity arises, we would like to work with NASA to empower Robonaut to do cognitive control. Applications 1. SES for Robonaut, 2. Robonaut Fault Diagnostic System, 3. ISAC Behavior Generation and Learning, 4. Segway Research.

Peters, R. A., II↗
NASA NTRS2020

Battery Health Quantification for TDRS Spacecraft by Using Signature Discriminability Measurement

The NASA/GSFC Space Network Project Office (SN) currently operates a constellation of ten geosynchronous TDRS spacecraft launched over the past 30 years. The SN project collects up to 16.5 Gigabytes of telemetry every month. Generally, the spacecraft health and functionality are obtained by the use of real-time telemetry data for the multiple spacecraft subsystems, which are transmitted to the main ground station at the White Sands Complex in Las Cruces, NM. Recently, the SN has instituted a program of Big Data to analyze the large amounts of data using a variety of tools including Machine Learning, Artificial Intelligence, development of training sets, and a variety of mathematical modeling tools. The goal is to improve spacecraft management and obtain a more accurate prediction of the spacecraft end of life. The combination of these efforts with those of the Aerospace Corporation, which has a contract with the SN to produce yearly reliability estimates for the TDRS fleet, will be performed. This paper presents a new concept called telemetry quality quantification (TQQ) and discusses the progress that has been made in battery performance estimation for the second-generation TDRS spacecraft using a signature discriminability measures (SDM) algorithm combined with the Aerospace Corp. battery life estimation models. This activity is important because many of the TDRS fleet of spacecraft have exceeded their on-orbit design lifetime and, therefore, NASA must carefully manage the spacecraft to continue operations while avoiding an end-of-mission scenario that leaves a non-functioning spacecraft in geosynchronous orbit.

Ma, Kenneth Y.↗
DOE OSTI2024

Engagement: Hyperparameter Optimization of Generative Adversarial Network Models for High-Energy Physics Simulations

We present our SciDAC FASTMath-HEP partnership results for tuning generative adversarial models (GANs) for high energy physics applications. The GANs are used in hybrid simulations to accelerate otherwise time-consuming computations. We optimize for both, prediction accuracy and variability with the goal to find GAN architectures that are reliable and robust.

high energy physics↗
DOE OSTI2026

matsim-agents v1.0

matsim-agents is a multi-agent AI framework for atomistic materials simulation and discovery. It orchestrates large language models (LLMs), machine-learned interatomic potentials (MLIPs), and DFT codes into a single agentic loop running on laptops and DOE leadership-class supercomputers. MULTI-AGENT ORCHESTRATION A LangGraph state machine with three nodes: a Planner that converts a natural-language research objective into structured tasks; an Executor that dispatches atomistic tools and loops until the queue is empty; and an Analyst that summarizes results into a human-readable report. State is checkpointed after every step and human-in-the-loop gates can be inserted at any edge. HYPOTHESIS-DRIVEN DISCOVERY CHAT An interactive REPL (matsim-agents chat) that couples LLM dialogue with atomistic simulation. Chemical formulas are automatically detected in conversation turns and trigger a full crystal-phase exploration: structure generation → relaxation → stability scoring → result injection back into the conversation, creating a closed hypothesis-refinement loop. CRYSTAL PHASE ENUMERATION Given a composition, the phase explorer enumerates prototypes by stoichiometry: elemental (fcc/bcc/hcp/sc/diamond), binary 1:1 (rocksalt/CsCl/zincblende/ wurtzite/fluorite/rutile), ternary 1:1:3 (cubic perovskite), ternary 1:2:4 (perovskite + spinel), quaternary 1:1:2:6 (Fm-3m double perovskite). 2-D prototypes (graphene, h-BN, MoS2 2H/1T) and multilayer stacking are also supported via --include-2d and --num-layers. SUPERCELL GENERATION AND SITE DECORATION Auto-tiling to a minimum atom count (--min-atoms), explicit NxNxN tiling (--supercell), symmetry-distinct site decorations (--n-orderings), and isotropic lattice-scale sweeps (--lattice-scales) for volume bracketing. MLFF RELAXATION AND STABILITY SCORING HydraGNN (multi-headed GNN) drives structure relaxation via ASE with FIRE, BFGS, or BFGSLineSearch. Stability output: delta-E/atom ranking across phases and a max-residual-force dynamical-stability proxy. Other MLIPs (MACE, NequIP, Orb) can be plugged in through the same interface. DFT BACKENDS Quantum ESPRESSO pw.x and VASP 6.6 are first-class labellers. Both have validated GPU builds and SLURM/PBS launchers for three DOE platforms: Frontier (AMD MI250X, ROCm), Aurora (Intel PVC, oneAPI), Perlmutter (NVIDIA A100, CUDA). QE produces ~100 binaries (pw.x, ph.x, epw.x, ...). VASP supports scf, relax, vc-relax, and vc-relax-shape run types. ACTIVE-LEARNING LOOP matsim-agents al run CONFIG.yaml drives an iterative HydraGNN-DFT loop: MD generates candidates → ensemble/MC-dropout uncertainty selects the most informative → DFT labels them in parallel inside one allocation → dataset grows → HydraGNN retrains → repeat. DFT backend is a single YAML toggle (dft.backend: vasp | qe). LLM-generated seed structures are supported (no curated POSCAR library needed). Config uses ${VAR}, ${VAR:-default}, ${VAR:?msg} shell-style substitution for cross-user/cross-site portability. LLM BACKENDS Ollama (local, default), vLLM (HPC multi-GPU serving), OpenAI, Anthropic, HuggingFace Transformers+Accelerate. Selected at runtime via flag or env var with no code changes. HPC PORTABILITY Same Python entry points run on Frontier (ROCm 7.2), Aurora (oneAPI), and Perlmutter (CUDA 12). DFT and ML stacks are never co-loaded in the same shell; they couple through the scheduler and filesystem. Advanced multi-node launchers (serve, discovery-chat, single-relaxation, active-learning, QE warm-start) are provided for all three platforms. CODABENCH COMPETITION BUNDLE A self-contained benchmark: 159 atomistic test structures across 11 material classes, 5 tasks (formation energy, forces, ML relaxation, AI-DFT relaxation, phase stability ranking), public/private leaderboard split (30/70), and four ready-to-run baselines: MACE-MP-0, HydraGNN, UMA, AllScAIP.

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗
DOE OSTI2026

Critical review of lithium recovery from geothermal brines with implications for Smackover Formation, USA

The rapidly growing demand for lithium, a critical element for energy storage and national security technologies, has intensified concerns over the long-term availability and environmental impact of conventional lithium sources, such as hard-rock mining. To meet future demand, it is vital to explore unconventional resources that can provide sustainable domestic supplies. Geothermal brines, produced as a byproduct of geothermal energy generation, offer a promising alternative for lithium recovery by leveraging existing infrastructure and renewable energy production. In particular, the Smackover Formation, an extensive reservoir of high-salinity brines spanning Arkansas, Texas, Louisiana, Mississippi, and Alabama in the U.S. Gulf Coast, holds significant untapped lithium reserves. Co-producing geothermal energy and lithium from these brines aligns with sustainable extraction objectives while addressing resource scarcity. This review synthesizes current knowledge of lithium occurrence in the Smackover Formation and geothermal resources in the region, while also exploring how emerging tools such as machine learning can enhance resource targeting and co-production efficiency. Finally, we discuss key technical challenges and outline future research directions needed to advance lithium extraction from geothermal brines and secure a resilient domestic supply chain.

15 GEOTHERMAL ENERGY↗
DOE OSTI2025

Real Time implementation of Artificial Intelligence compression algorithm for High-Speed Streaming Readout signals

The new generation of high-energy physics experiments plans to acquire data in streaming mode. With this approach, it is possible to access the information of the whole detector (organized in time slices) for optimal and lossless triggering of data acquisitions. With this approach, data rates, especially in large detectors, are often very high, and the network is likely to be the bottleneck for the entire Streaming Read Out system. The aim of this work is to study the implementation of a lossy compression algorithm based on Artificial Intelligence: an Autoencoder. With Machine Learning it is possible to achieve a high compression ratio and fast inference time with only a small degradation of the signals, almost negligible for the specific application. This work explores different configurations of the Autoencoder and the implementation on different hardware. Different Autoencoder configurations are explored to find the best trade-off between compression ratio and reconstruction loss, both for signals and energy spectrum. Different hardware implementations are also explored to find the best platform to achieve real-time performance for the specific application.

Rossi, Fabio (ORCID:0009000385713885)↗
DOE OSTI2026

Macroscopic trends of neoclassical tearing stability in high-field H-mode tokamak pilot plants

The neoclassical tearing mode (NTM) stability metric—minimum marginally stable island width $w$$^{*}_{m}$—was compared across 14651 inductive high-field tokamak pilot plant equilibria. Larger devices with reduced elongation and/or increased minor radius demonstrated an order-of-magnitude increase in $w$$^{*}_{m}$, primarily due to a reduction in bootstrap drive. This work is part of an ongoing effort to ensure passive NTM-stability in the ARC tokamak, in which the technology to achieve active tearing-suppression with localised electron cyclotron current drive does not yet exist. The equilibrium scenarios in the database were Monte Carlo generated and normalised to the same >400MW fusion power, minimum pressure scenario at a range of plasma currents, before tearing analysis using the modified Rutherford equation was applied for all resonant poloidal and toroidal m, n modes up to n = 4. Single-helicity toroidal Δ' calculations in resistive DCON set the minimum marginally stable island width, and a simple modal scaling proportional to –m 2 n –1 was identified for high-m Δ' values. The dominant correlates of $w$$^{*}_{m}$ and Δ' across the database were analysed using interpretable machine learning techniques.

NTM seeding↗
DOE OSTI2024

Prediction of the Cu oxidation state from EELS and XAS spectra using supervised machine learning

Abstract Electron energy loss spectroscopy (EELS) and X-ray absorption spectroscopy (XAS) provide detailed information about bonding, distributions and locations of atoms, and their coordination numbers and oxidation states. However, analysis of XAS/EELS data often relies on matching an unknown experimental sample to a series of simulated or experimental standard samples. This limits analysis throughput and the ability to extract quantitative information from a sample. In this work, we have trained a random forest model capable of predicting the oxidation state of copper based on its L-edge spectrum. Our model attains an R 2 score of 0.85 and a root mean square error of 0.24 on simulated data. It has also successfully predicted experimental L-edge EELS spectra taken in this work and XAS spectra extracted from the literature. We further demonstrate the utility of this model by predicting simulated and experimental spectra of mixed valence samples generated by this work. This model can be integrated into a real-time EELS/XAS analysis pipeline on mixtures of copper-containing materials of unknown composition and oxidation state. By expanding the training data, this methodology can be extended to data-driven spectral analysis of a broad range of materials.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Digital Twin User Guide for Chelan County Public Utility District

This user manual offers a comprehensive guide for developing a Digital twin (DT) of a Kaplan turbine at Chelan County Public Utility District (Chelan PUD) using neural networks. As variable renewable generation expands, hydropower units must operate with optimal efficiency and stability. For Kaplan machines, this flexibility is achieved through coordinated control of guide vane (wicket gates) opening and runner blade pitch, which amplifies the plant’s inherent nonlinear behavior and challenges traditional physics-only modeling. The efficiency of the Kaplan turbine varies with different combinations of the guide vans (wicket gate) opening and the blade angle. Each guide van opening and blade angle has a corresponding highest efficiency point, forming a cam relationship that represents the optimal combination.The discharge of a hydraulic turbine is controlled by the opening angle of the guide vans. Therefore, for each value of head, there is a certain guide van opening and blade angle that corresponds to the highest efficiency. For a given head, different combinations of the guide van opening and blade angle have different efficiencies. Therefore, coordinate cam curves are used to describe the relationship between the wicket gate opening and blade angle with different water head. To address these challenges, the manual details a data-driven modeling and learning workflow centered on structured neural networks. The approach is designed to forecast critical operational variables—discharge flow, net head, penstock (or scroll-case) pressure, and generator electrical outputs—by leveraging real-time inputs such as the generator power control setpoint, exciter field current and field voltage, together with hydromechanical commands (e.g., gate position and, when available, runner blade-pitch angle). The neural models are trained and validated on operational data from a Kaplan unit operated by Chelan PUD, demonstrating that the structured NN architecture can learn the coupled gate–blade–electrical dynamics. The result is a robust DT that improves situational awareness and supports data-informed decision-making for Chelan PUD’s Kaplan turbine operations.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Dynamics and lipid membrane coupling of the RAS-RAF complex revealed via multiscale simulations

To gain molecular and mechanistic insights into initiation of the RAS-RAF signaling cascade, we developed and used a combination of multiscale simulation and experimental approaches. The influence and impact of the membrane on RAS and RAF proteins is a factor we are just beginning to understand and appreciate in more detail. Molecular simulation is an ideal methodology to further study this complicated relationship between the membrane and associated proteins. Our previous work using Multiscale Machine-learned Modeling Infrastructure investigated different lipid compositions solely around the KRAS4b protein and the interplay between protein behavior and these membrane environments. Multiscale Machine-learned Modeling Infrastructure uses machine learning to couple adjacent simulation scales and has been efficiently scaled across some of the world’s largest high-performance computers. Recently, we have expanded this multiresolution framework to include the all-atom simulation scale and to incorporate the RAF RBDCRD domains. Here, we present the overall analysis results from this new simulation campaign comprising a mixture of RAS and RAF RBDCRD proteins. Approximately 35,000 coarse-grained and 10,000 all-atom molecular dynamics simulations were completed, sampled from a variety of protein/lipid composition configurations that were generated from a micron-scale continuum simulation containing hundreds of copies of the proteins. Our studies suggest that orientations of the RAS-RBDCRD complex on the membrane occupy distinct configurational states, and the spatial patterns of lipid arrangements around these different protein states are unique to each state. The extent and size of lipid “fingerprints” imposed on the membrane by the RAS-RBDCRD protein complex are significantly larger than observed for just the RAS protein on its own. These protein complexes strongly associate, but we do not observe statistically significant preferred protein-protein orientations. These observations indicate that spatial colocalization of RAS-RBDCRD proteins in the same vicinity may be assisted by specific membrane environments, acting to increase the probability of signaling complex formation.

Carpenter, Timothy S. [Lawrence Livermore National↗
NASA NTRSDate not supplied

Development of Carbon Flux Model Using ABI Data Over the Conterminous US

The satellite-driven carbon flux estimation has been playing important role to estimate continental-scale carbon budget. One of the biggest recent advances in the satellite-driven carbon flux modeling is utilization of high-frequent geostationary satellites to estimate diurnal cycle in carbon fluxes. The satellite based carbon flux estimation used the polar orbiting satellite sensors (e.g., Moderate Resolution Imaging Spectroradiometer (MODIS)), which allow us to observe target regions only once during the day. The new generation of geostationary satellite sensors provide frequent observations, often less than every 10 minutes. Here, we use GOES Advanced Baseline Imager (ABI) data to estimate hourly NEE over the conterminous US. We used the Terrestrial Observation Prediction System (TOPS) model for estimating hourly NEE. TOPS is a diagnostic ecosystem process model that simulates the fluxes of carbon and water through vegetation in response to climate variability. For the climate input, we developed hourly climate data using the same algorithm with NASA Earth Exchange Gridded Daily Meteorology (NEX-GDM) datasets based on machine learning techniques. The hourly climate data includes precipitation, maximum temperature, minimum temperature, dew point temperature, and solar radiation were derived from the Geostationary observations. The spatial patterns of ecosystem parameters used in TOPS are optimized using satellite Solar Induced Fluorescence (SIF) data. The high frequency GPP estimations from geostationary satellite sensors make it comparable to the instantaneous SIF data than daily GPP. We also used Ameriflux data for optimization of model parameters and the validation of the output. The derived data addresses the diurnal dynamics of carbon cycling at large scales and should help in reducing the uncertainties in carbon budget studies.

geostationary satellite↗
DOE OSTI2025

Resimulation-based self-supervised learning for pretraining physics foundation models

Self-supervised learning (SSL) is at the core of training modern large machine learning models, providing a scheme for learning powerful representations that can be used in a variety of downstream tasks. However, SSL strategies must be adapted to the type of training data and downstream tasks required. We propose resimulation-based self-supervised representation learning (RS3L), a novel simulation-based SSL strategy that employs a method of resimulation to drive data augmentation for contrastive learning in the physical sciences, particularly, in fields that rely on stochastic simulators. By intervening in the middle of the simulation process and rerunning simulation components downstream of the intervention, we generate multiple realizations of an event, thus producing a set of augmentations covering all physics-driven variations available in the simulator. Using experiments from high-energy physics, we explore how this strategy may enable the development of a foundation model; we show how RS3L pretraining enables powerful performance in downstream tasks such as discrimination of a variety of objects and uncertainty mitigation. In addition to our results, we make the RS3L dataset publicly available for further studies on how to improve SSL strategies.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Squeezing Every Last 'Bit' of Information from Enceladus Mass Spectrometry

Potential opportunities to return to Enceladus in Discovery and Flagship class missions inspire development of next-generation instruments and creative approaches to sample collection, sample analysis, and data analysis and transmission strategies. Mass spectrometers (MS) are ideally suited to future Enceladus missions due to their analytical power in identifying a range of molecular and ionic compositions – including complex organics – and potentially astrobiologically-important features such as isotope ratios, chirality, and enantiomeric excess. However, long communication delays from Enceladus and limited bandwidth limits the data transmission from these higher-data-volume instruments, likely delaying mission-related response to new data. We explore the utility of data science and machine learning (ML) on isotope ratio (IR)MS data collected from laboratory analogs of Enceladus to: 1) process data quickly for rapid ground-based analyses, 2) understand if compositional and biosignature information could be extracted from IRMS data, and 3) evaluate whether onboard ML techniques could improve sample analysis, cadence, and transmission prioritization. Laboratory analogs analyzed isotopes of volatile CO2 that interacted with seawaters of varying composition, and include both abiotic and biotic (microbially-influenced) experiments. Enceladus’s alkaline oceans promote speciation of carbon into multiple forms (e.g., H2CO3 / CO2, HCO3-, and CO32-), each of which could be isotopically fractionated by abiotic or biotic reactions. Large (>2‰) changes in carbon isotopes (δ13C) are observed from some biotic experiments inoculated with complex microbial ecosystems relative to the abiotic seawaters. ML training and classification suggests that microbial samples can be distinguished from abiotic samples, yet that a broad range of microbial experiments are necessary to train ML models to cover a range of complexities including disequilibria, and isotopic and compositional fractionation.

geochemistry↗
DOE OSTI2024

Non-Electricity Based Renewable Fuels: Theory and Computation for Solar Thermochemical Hydrogen

Dominated by photovoltaics and wind, current renewable energy sources generate mostly electricity, but 80% of the global final energy consumption occurs in form of fuels. Therefore, direct solar fuel generation would be a major breakthrough for the energy transition. Solar thermochemical hydrogen (STCH) is one of the very few potential routes towards scalable renewable fuels, but currently suffers from lack of an oxide working material that could optimally perform energy conversion within the thermodynamic boundary conditions. Theory and computation can contribute in two distinct ways, through materials search and discovery, but also by providing detailed mechanistic models for specific systems so to advance our understanding of possible design strategies. To enable high-throughput materials screening, we developed a defect graph neural network (dGNN) machine learning approach,[1] which accelerates the prediction of defect formation energies by replacing the tedious density functional theory (DFT) supercell calculations for all possible defect sites. This approach enables high-throughput database screening of oxides, which was integrated with thermodynamic modeling to extract the reduction entropies as additional selection criterion for STCH. Once potential candidate materials are identified, detailed models can guide materials design by predicting performance characteristics. One challenge is to quantitatively predict thermochemical equilibria at high concentrations when the redox active defects start to interact with each other, thereby impeding the formation of additional defects. Introducing a model for the free energy of defect interaction, parametrized on the basis of DFT data, we simulated the complete STCH redox cycle for (Sr,Ce)MnO3 alloys, achieving near-quantitative agreement with experimental data.[2] The analysis of these simulations reveals how defect interactions diminish the reduction entropy and H2 yield, suggesting to include these interactions in design considerations. Finally, we revisit the popular van't Hoff method for analyzing reduction enthalpies and entropies. This method is not ideal, as it involves a temperature-dependent convolution of gas-phase and solid-state entropies, causing uncertainties in the same order of magnitude as the physical quantities of interest. To avoid this problem, we suggest a simple alternative approach which can be applied to experimental and simulated data alike.

first-principles calculations↗
DOE OSTI2024

Automating Rabi & Ramsey Measurements via Machine Learning

As quantum computers scale up, the manual process of qubit tune-up becomes increasingly impractical due to its time-consuming and repetitive nature. While existing research has explored some automation techniques, many models remain underutilized for this purpose. This research aims to answer the question: can qubit tune-up be automated using the Long Short-Term Memory (LSTM) model? For the purposes of this project, only the rabi and ramsey measurement cycle was automated. These measurements are used to fine-tune a rough qubit frequency by repeating them until the optimal qubit frequency is obtained. The LSTM model uses the qubit frequency at one time step to forecast the qubit frequency at the next time step. A rabi-ramsey simulation was made to fabricate a dataset to train and test the LSTM model. As the model was trained, the error of the model decreased. Although there wasn't enough training data to generate perfect predictions, this shows it is possible to utilize forecasting models in automating the tune-up process.

Roberts, Rachel↗
DOE OSTI2025

SymbolFit: Automatic Parametric Modeling with Symbolic Regression

We introduce SymbolFit (API: https://github.com/hftsoi/symbolfit), a framework that automates parametric modeling by using symbolic regression to perform a machine-search for functions that fit the data while simultaneously providing uncertainty estimates in a single run. Traditionally, constructing a parametric model to accurately describe binned data has been a manual and iterative process, requiring an adequate functional form to be determined before the fit can be performed. The main challenge arises when the appropriate functional forms cannot be derived from first principles, especially when there is no underlying true closed-form function for the distribution. In this work, we develop a framework that automates and streamlines the process by utilizing symbolic regression, a machine learning technique that explores a vast space of candidate functions without requiring a predefined functional form because the functional form itself is treated as a trainable parameter, making the process far more efficient and effortless than traditional regression methods. We demonstrate the framework in high-energy physics experiments at the CERN Large Hadron Collider (LHC) using five real proton-proton collision datasets from new physics searches, including background modeling in resonance searches for high-mass dijet, trijet, paired-dijet, diphoton, and dimuon events. We show that our framework can flexibly and efficiently generate a wide range of candidate functions that fit a nontrivial distribution well using a simple fit configuration that varies only by random seed, and that the same fit configuration, which defines a vast function space, can also be applied to distributions of different shapes, whereas achieving a comparable result with traditional methods would have required extensive manual effort.

Tsoi, Ho Fung [Univ. of Pennsylvania, Philadelphia↗
DOE OSTI2025

Learning the simplicity of scattering amplitudes

The simplification and reorganization of complex expressions lies at the core of scientific progress, particularly in theoretical high-energy physics. This work explores the application of machine learning to a particular facet of this challenge: the task of simplifying scattering amplitudes expressed in terms of spinor-helicity variables. We demonstrate that an encoder-decoder transformer architecture achieves impressive simplification capabilities for expressions composed of handfuls of terms. Lengthier expressions are implemented in an additional embedding network, trained using contrastive learning, which isolates subexpressions that are more likely to simplify. The resulting framework is capable of reducing expressions with hundreds of terms—a regular occurrence in quantum field theory calculations—to vastly simpler equivalent expressions. Starting from lengthy input expressions, our networks can generate the Parke-Taylor formula for five-point gluon scattering, as well as new compact expressions for five-point amplitudes involving scalars and gravitons.

Cheung, Clifford [California Institute of Technolo↗