Search NASA⌕ Search

SEARCH · Search NASA

Results for “Deep generative model”

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 793 records · Page 44

AI‐Driven Defect Engineering for Advanced Thermoelectric Materials

Thermoelectric materials offer a promising pathway to directly convert waste heat to electricity. However, achieving high performance remains challenging due to intrinsic trade-offs between electrical conductivity, the Seebeck coefficient, and thermal conductivity, which are further complicated by the presence of defects. This review explores how artificial intelligence (AI) and machine learning (ML) are transforming thermoelectric materials design. Advanced ML approaches including deep neural networks, graph-based models, and transformer architectures, integrated with high-throughput simulations and growing databases, effectively capture structure-property relationships in a complex multiscale defect space and overcome the “curse of dimensionality”. This review discusses AI-enhanced defect engineering strategies such as composition optimization, entropy and dislocation engineering, and grain boundary design, along with emerging inverse design techniques for generating materials with targeted properties. Finally, it outlines future opportunities in novel physics mechanisms and sustainability, highlighting the critical role of AI in accelerating the discovery of thermoelectric materials.

36 MATERIALS SCIENCE↗

Unveiling mesoscopic structures in distorted lamellar phases through deep learning-based small angle neutron scattering analysis

Hypothesis: The formation of distorted lamellar phases, distinguished by their arrangement of crumpled, stacked layers, is frequently accompanied by the disruption of long-range order, leading to the formation of interconnected network structures commonly observed in the sponge phase. Nevertheless, traditional scattering functions grounded in deterministic modeling fall short of fully representing these intricate structural characteristics. Our hypothesis posits that a deep learning method, in conjunction with the generalized leveled wave approach used for describing structural features of distorted lamellar phases, can quantitatively unveil the inherent spatial correlations within these phases. Experiments and Simulations: This report outlines a novel strategy that integrates convolutional neural networks and variational autoencoders, supported by stochastically generated density fluctuations, into a regression analysis framework for extracting structural features of distorted lamellar phases from small angle neutron scattering data. To evaluate the efficacy of our proposed approach, we conducted computational accuracy assessments and applied it to the analysis of experimentally measured small angle neutron scattering spectra of AOT surfactant solutions, a frequently studied lamellar system. Findings: The findings unambiguously demonstrate that deep learning provides a dependable and quantitative approach for investigating the morphology of wide variations of distorted lamellar phases. It is adaptable for deciphering structures from the lamellar to sponge phase including intermediate structures exhibiting fused topological features. In conclusion, this research highlights the effectiveness of deep learning methods in tackling complex issues in the field of soft matter structural analysis and beyond.

36 MATERIALS SCIENCE↗

Interpretable Deep Learning for Advancing Field-Enhanced Catalysis

This DOE Early Career project developed a physics-informed, interpretable AI-and-modeling framework to understand and exploit electric-field effects in heterogeneous catalysis, with ammonia cracking and synthesis as a representative pathway. The team built and validated methods to map local electric fields on metal surfaces and nanoparticles, showing that low-coordination features (tips/edges/corners) can concentrate fields by several-fold relative to flat facets. Using DFT-generated datasets, the project created physics-guided machine learning models that rapidly predict local electric fields and field-dependent adsorption energetics with near-DFT accuracy while reducing computational cost by orders of magnitude. These predictions were integrated with microkinetic modeling to quantify how field-dipole interactions reshape reaction energetics and mechanisms, enabling large increases in predicted catalytic rates and substantial reductions in operating temperature under favorable field conditions. To accelerate discovery of earth-abundant catalysts, the project combined interpretable ML screening (with electronic-structure descriptors identified as key drivers) with a generative inverse-design workflow based on diffusion models and physics constraints. The resulting closed-loop approach, linking simulation, mechanistic modeling, and AI, provides reusable tools and datasets for designing catalysts and operating conditions in field-enhanced catalysis, with broad relevance to electrostatic catalysis, plasma catalysis, electrocatalysis, and other energy-related chemical transformations.

30 DIRECT ENERGY CONVERSION↗

Recent Developments in SHERPA

Some recent QCD-related developments in the SHERPA event generator are presented. In the past decades, event generators such as PYTHIA [1, 2] and HERWIG [3, 4] have been central for nearly all physics analyses at particle physics experiments at the high-energy frontier. This will also hold true at the LHC, where a large number of interesting signals for new particles or new phenomena (the Higgs boson or any other manifestation of the mechanism behind electro-weak symmetry breaking, supersymmetry, extra dimensions etc.) is hampered by a plethora of severe, sometimes overwhelming backgrounds. Nearly all of them are largely influenced by QCD. Therefore it seems fair to say that the success of the LHC in finding new physics may very well depend on a deep and detailed understanding of old physics, like QCD. Examples for this include, among others, the central-jet veto for the vector boson fusion channel for Higgs production or topologies, where gauge bosons emerge in association with many jets, a background for many search channels. In a reflection on increased needs by the experimental community, aiming at higher precision, incorporation of new physics models and so on, the work horses of old have undergone serious renovation efforts, resulting in new, improved versions of the respective codes, namely PYTHIA8 [5] and HERWIG++ [6]. In addition a completely new code, SHERPA [7], has been constructed and is in the process of maturing. The status of this code is the topic of this contribution. SHERPA's hallmark property is the inclusion of higher-order tree-level QCD contributions, leading to an improved modelling of jet production. They are introduced through a full-fledged matrix element generator, AMEGIC++ [8], which is capable of generating matrix elements and corresponding phase space mappings for processes with multi-particle final states in various models, including the Standard Model, anomalous gauge triple and quadruple couplings according to [9, 10], the Minimal Supersymmetric Standard Model with Feynman rules from [11], the ADD-model of extra dimensions [12, 13], and a model with an extra U(1) singlet coupling to the Higgs boson only [14]. The code has been thoroughly tested and validated [15]. This code, however, is limited, especially in the treatment of many ({ge} 6) external QCD particles. Therefore, in the near future, SHERPA will incorporate another, new matrix element generator, COMIX, which is based on Berends-Giele recursion relations [16] and color-dressing [17] rather than color-ordering. In Tabs. 1 and 2 some example cross sections for gg {yields} ng at fixed energies and pp {yields} b{bar b} + n jets obtained with this program are exhibited and compared to those from other programs. In addition, concerning the calculation of higher-order matrix elements and cross sections, there have been first steps towards an automation of such calculations at truly next-to leading order accuracy. They manifest themselves in the implementation of a procedure [19] to fully automatically construct and evaluate Catani-Seymour dipole subtraction terms [20] for the real part of such NLO calculations. The results from the matrix element calculations are merged with the subsequent parton shower through the formalism of [21, 22]. The results of its implementation in SHERPA [23] has recently been compared with other algorithms [24]. Although there remains some dispute about the theoretical equivalence of the different approaches, the overall results show satisfying agreement with each other, such that they can be used with confidence for data analysis.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Vertex finding in neutrino-nucleus interaction: a model architecture comparison

We compare different neural network architectures for machine learning algorithms designed to identify the neutrino interaction vertex position in the MINERvA detector. The architectures developed and optimized by hand are compared with the architectures developed in an automated way using the package “Multi-node Evolutionary Neural Networks for Deep Learning” (MENNDL), developed at Oak Ridge National Laboratory. While the domain-expert hand-tuned network was the best performer, the differences were negligible and the auto-generated networks performed as well. There is always a trade-off between human, and computer resources for network optimization and this work suggests that automated optimization, assuming resources are available, provides a compelling way to save significant expert time.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach

Wind speed prediction with spatio–temporal correlation is among the most challenging tasks in wind speed prediction. In this paper, the problem of predicting wind speed for multiple sites simultaneously is investigated by using spatio–temporal correlation. This paper proposes a model for wind speed prediction with spatio–temporal correlation, i.e., the predictive deep convolutional neural network (PDCNN). The model is a unified framework, integrating convolutional neural networks (CNNs) and a multi-layer perceptron (MLP). Firstly, the spatial features are extracted by CNNs located at the bottom of the model. Then, the temporal dependencies among these extracted spatial features are captured by the MLP. In this way, the spatial and temporal correlations are captured by PDCNN intrinsically. Finally, PDCNN generates the predicted wind speed by using the learnt spatio–temporal correlations. In addition, three error indices are defined to evaluate the prediction accuracy of the model on the wind turbine array. Experiment results on real-world data show that PDCNN can capture the spatio–temporal correlation effectively, and it outperforms the conventional machine learning models, including multi-layer perceptron, support vector regressor, decision tree, etc.

17 WIND ENERGY↗

Considering the Role of Natural Gas in the Deep Decarbonization of the U.S. Electricity Sector. Natural Gas and the Evolving U.S. Power Sector Monograph Series: Number 2

Natural gas generation in the U.S. electricity sector has grown substantially in recent years, while the sector's carbon dioxide (CO2) emissions have generally declined. This relationship highlights the concept of natural gas as a potential enabler of a transition to a lower-carbon future. This work considers that concept by using the National Renewable Energy Laboratory (NREL) Renewable Energy Deployment System (ReEDS) model. ReEDS is a long-term capacity expansion model of the U.S. electricity sector. We examine the role of natural gas within the ReEDS modeling framework as increasingly strict carbon emission targets are imposed on the electricity sector. In addition to various natural gas price futures, we also consider scenarios that emphasize a low-carbon technology in order to better understand the role of natural gas if that low-carbon technology shows particular promise. Specifically, we consider scenarios with high amounts of energy efficiency (EE), low nuclear power costs, low renewable energy (RE) costs, and low carbon capture and storage (CCS) costs. Within these scenarios we find that requiring the electricity sector to lower CO2 emissions over time increases near-to-mid-term (through 2030) natural gas generation (see Figure 1 - left). The long-term (2050) role of natural gas generation in the electricity sector is dependent on the level of CO2 emission reduction required. Moderate reductions in long-term CO2 emissions have relatively little impact on long-term natural gas generation, while more stringent CO2 emission limits lower long-term natural gas generation (see Figure 1 - right). More stringent carbon targets also impact other generating technologies, with the scenarios considered here seeing significant decreases in coal generation, and new capacity of nuclear and renewable energy technologies over time. Figure 1 also demonstrates the role of natural gas in the context of scenarios where a specific low-carbon technology is advantaged. In 2030, natural gas generation in the technology scenarios is quite similar to that in the reference scenarios, indicating relatively little change in the role of natural gas in the near-to-mid-term due to advancements in those technology areas. The 2050 natural gas generation shows more significant differences, suggesting that technology advancements will likely have substantial impacts on the role of natural gas in the longer-term timeframe. Natural gas generation differences are most strongly driven by alternative natural gas price trajectories--changes in natural gas generation in the Low NG Price and High NG Price scenarios are much larger than in any other scenario in both the 2030 and 2050 timeframes. The only low-carbon technology scenarios that showed any increase in long-term natural gas generation relative to the reference case were the Low CCS cost scenarios. Carbon capture and storage technology costs are currently high, but have the potential to allow fossil fuels to play a larger role in low-carbon grid. This work considers three CCS cost trajectories for natural gas and coal generators: a baseline trajectory and two lower cost trajectories where CO2 capture costs reach $40/metric ton and $10/metric ton, respectively. We find that in the context of the ReEDS model and with these assumed cost trajectories, CCS can increase the long-term natural gas generation under a low carbon target (see Figure 2). Under less stringent carbon targets we do not see ReEDS electing to use CCS as part of its electricity generating portfolio for the scenarios considered in this work.

03 NATURAL GAS↗

Cloud, Aerosol, and Complex Terrain Interactions (CACTI) Science Plan

General circulation models and downscaled regional models exhibit persistent biases in deep convective initiation location and timing, cloud top height, stratiform area and precipitation fraction, and anvil coverage. Despite important impacts on the distribution of atmospheric heating, moistening, and momentum, nearly all climate models fail to represent convective organization, while system evolution is not represented at all. Improving representation of convective systems in models requires characterization of their predictability as a function of environmental conditions, and this characterization depends on observing many cases of convective initiation, non-initiation, organization, and non-organization. The Cloud, Aerosol, and Complex Terrain Interactions (CACTI) experiment in the Sierras de Córdoba mountain range of north-central Argentina is designed to improve understanding of cloud life cycle and organization in relation to environmental conditions so that cumulus, microphysics, and aerosol parameterizations in multiscale models can be improved. The Sierras de Córdoba range has a high frequency of orographic boundary-layer clouds, many reaching congestus depths, many initiating into deep convection, and some organizing into mesoscale systems uniquely observable from a single fixed site. Some systems even grow upscale to become among the deepest, largest, and longest-lived in the world. These systems likely contribute to an observed regional trend of increasing extreme rainfall, and poor prediction of them likely contributes to a warm, dry bias in climate models downstream of the Sierras de Córdoba range in a key agricultural region. Many environmental factors influence the convective life cycle in this region, including orographic, low-level jet, and frontal circulations, surface fluxes, synoptic vertical motions influenced by the Andes, cloud detrainment, and aerosol properties. Local and long-range transport of smoke resulting from biomass burning as well as blowing dust are common in the austral spring, while changes in land surface properties as the wet season progresses impact surface fluxes and boundary-layer evolution on daily and seasonal time scales that feed back to cloud and rainfall generation. This range of environmental conditions and cloud properties coupled with a high frequency of events makes this an ideal location for improving our understanding of cloud-environment interactions. The following primary science questions will be addressed through coordinated first U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Mobile Facility (AMF1), mobile C-band Scanning ARM Precipitation Radar (C-SAPR2), ARM Aerial Facility (AAF) Gulfstream-1 (G-1), and guest instrument observations: How are the properties and life cycles of orographically generated cumulus humulis, mediocris, and congestus clouds affected by environmental kinematics, thermodynamics, aerosols, and surface properties? How do these cloud types alter these environmental conditions? How do environmental kinematics, thermodynamics, and aerosols impact deep convective initiation, upscale growth, and mesoscale organization? How are soil moisture, surface fluxes, and aerosol properties altered by deep convective precipitation events and seasonal accumulation of precipitation? This multi-faceted experiment involves a long-term 7-month extended operational period (EOP, 1 October, 2018–30 April, 2019) as well as a 1.5-month intensive operational period (IOP, 30 October–13 December) that will include G-1 observations coinciding with the international multi-agency Remote Sensing of Electrification, Lightning, and Meso-scale/Micro-scale Processes with Adaptive Ground Observations (RELAMPAGO) field campaign.

Convection, Andes, Argentina, cloud life cycle, cl↗

Pilot Residential Deep Energy Retrofits and the PNNL Lab Homes

This report summarizes research investigating the technical and economic feasibility of several pilot deep energy retrofits, or retrofits that save 30% to 50% or more on a whole-house basis while increasing comfort, durability, combustion safety, and indoor air quality. The work is being conducted for the U.S. Department of Energy Building Technologies Program as part of the Building America Program. As part of the overall program, Pacific Northwest National Laboratory (PNNL) researchers are collecting and analyzing a comprehensive dataset that describes pre- and post-retrofit energy consumption, retrofit measure cost, health and comfort impacts, and other pertinent information for each home participating in the study. The research and data collection protocol includes recruitment of candidate residences, a thorough test-in audit, home energy modeling, and generation of retrofit measure recommendations, implementation of the measures, test-out, and continued evaluation. On some homes, more detailed data will be collected to disaggregate energy-consumption information. This multi-year effort began in October 2010. To date, the PNNL team has performed test-in audits on 51 homes in the marine, cold, and hot-humid climate zones, and completed 3 retrofits in Texas, 10 in Florida, and 2 in the Pacific Northwest. Two of the retrofits are anticipated to save 50% or more in energy bills and the others - savings are in the 30% to 40% range. Fourteen other retrofits are under way in the three climate zones. Metering equipment has been installed in seven of these retrofits - three in Texas, three in Florida, and one in the Pacific Northwest. This report is an interim update, providing information on the research protocol and status of the PNNL deep energy retrofit project as of December, 2011. The report also presents key findings and lessons learned, based on the body of work to date. In addition, the report summarizes the status of the PNNL Lab Homes that are new manufactured homes procured with minimal energy-efficiency specifications typical of existing homes in the region, and sited on the PNNL campus. The Lab Homes serve as a flexible test facility (the first of its kind in the Pacific Northwest) to rapidly evaluate energy-efficient and grid-smart technologies that are applicable to residential construction.

54 ENVIRONMENTAL SCIENCES↗

Introducing single-crystal scattering and optical potentials into MCNPX: Predicting neutron emission from a convoluted moderator

Neutron transport simulation codes are an indispensable tool used for the design and construction of modern neutron scattering facilities and instrumentation. It has become increasingly clear that some neutron instrumentation has started to exploit physics that is not well-modelled by the existing codes. Particularly, the transport of neutrons through single crystals and across interfaces in MCNP(X), Geant4 and other codes ignores scattering from oriented crystals and refractive effects, and yet these are essential ingredients for the performance of monochromators and ultra-cold neutron transport respectively (to mention but two examples). In light of these developments, we have extended the MCNPX code to include a single-crystal neutron scattering model and neutron reflection/refraction physics. Furthermore, we have also generated silicon scattering kernels for single crystals of definable orientation with respect to an incoming neutron beam. As a first test of these new tools, we have chosen to model the recently developed convoluted moderator concept, in which a moderating material is interleaved with layers of perfect crystals to provide an exit path for neutrons moderated to energies below the crystal s Bragg cut off at locations deep within the moderator. Studies of simple cylindrical convoluted moderator systems of 100 mm diameter and composed of polyethylene and single crystal silicon were performed with the upgraded MCNPX code and reproduced the magnitude of effects seen in experiments compared to homogeneous moderator systems. Applying different material properties for refraction and reflection, and by replacing the silicon in the models with voids, we show that the emission enhancements seen in recent experiments are primarily caused by the transparency of the silicon/void layers. Finally the convoluted moderator experiments described by Iverson et al. were simulated and we find satisfactory agreement between the measurement and the results of simulations performed using the tools we have developed.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

OHMIC DISSIPATION IN THE ATMOSPHERES OF HOT JUPITERS

Hot Jupiter atmospheres exhibit fast, weakly ionized winds. The interaction of these winds with the planetary magnetic field generates drag on the winds and leads to ohmic dissipation of the induced electric currents. We study the magnitude of ohmic dissipation in representative, three-dimensional atmospheric circulation models of the hot Jupiter HD 209458b. We find that ohmic dissipation can reach or exceed 1% of the stellar insolation power in the deepest atmospheric layers, in models with and without dragged winds. Such power, dissipated in the deep atmosphere, appears sufficient to slow down planetary contraction and explain the typically inflated radii of hot Jupiters. This atmospheric scenario does not require a top insulating layer or radial currents that penetrate deep in the planetary interior. Circulation in the deepest atmospheric layers may actually be driven by spatially non-uniform ohmic dissipation. A consistent treatment of magnetic drag and ohmic dissipation is required to further elucidate the consequences of magnetic effects for the atmospheres and the contracting interiors of hot Jupiters.

79 ASTRONOMY AND ASTROPHYSICS↗

$\mathrm{SageNet}$: Fast Neural Network Emulation of the Stiff-amplified Gravitational Waves from Inflation

Accurate modeling of the inflationary gravitational waves (GWs) requires time-consuming, iterative numerical integrations of differential equations to take into account their backreaction on the expansion history. To improve computational efficiency while preserving accuracy, we present the Stiff-amplified Gravitational-wave Emulator Network (SageNet), a deep learning framework designed to replace conventional numerical solvers (code available at https://github.com/YifangLuo/SageNet). SageNet employs a long short-term memory architecture to emulate the present-day energy density spectrum of the inflationary GWs with possible stiff amplification, Ω GW (f). Trained on a data set of 25,689 numerically generated solutions, SageNet allows accurate reconstructions of Ω GW (f) and generalizes well to a wide range of cosmological parameters; 90.9% of the test emulations with randomly distributed parameters exhibit errors of under 4%. In addition, SageNet demonstrates its ability to learn and reproduce the artificial, adaptive sampling patterns in numerical calculations, which implement denser sampling of frequencies around changes in spectral indices in Ω GW (f). The dual capability of learning both physical and artificial features of the numerical GW spectra establishes SageNet as a robust alternative to exact numerical methods. Finally, our benchmark tests show that SageNet reduces the computation time from tens of seconds to milliseconds, achieving a speedup of ∼10 4 times over standard CPU-based numerical solvers with the potential for further acceleration on GPU hardware. These capabilities make SageNet a powerful tool for accelerating Bayesian inference procedures for extended cosmological models. In a broad sense, the SageNet framework offers a fast, accurate, and generalizable solution to modeling cosmological observables whose theoretical predictions demand costly differential equation solvers.

Astronomy data modeling↗

Spatial Variability of the Background Diurnal Cycle of Deep Convection around the GoAmazon2014/5 Field Campaign Sites

The Amazon rainforest is one of a few regions of the world where continental tropical deep convection occurs. The Amazon’s isolation makes it challenging to observe, but also creates a unique natural laboratory to study anthropogenic impacts on clouds and precipitation in an otherwise pristine environment. Extensive measurements were made upwind and downwind of the large city of Manaus, Brazil during the Observations and Modeling of the Green Ocean Amazon 2014-2015 (GoAmazon2014/5) field campaign. In this study, 15 years of high-resolution satellite data are analyzed to examine the spatial and diurnal variability of convection occurring around the GoAmazon2014/5 sites. Interpretation of anthropogenic differences between the upwind (T0) and downwind (T1-T3) sites is complicated by naturally-occurring spatial variability between the sites. During the rainy season, the inland propagation of the previous day’s sea-breeze front happens to be in phase with the background diurnal cycle near Manaus, but is out of phase elsewhere. Enhanced convergence between the river-breezes and the easterly trade winds generates up to 10% more frequent deep convection at the GoAmazon2014/5 sites east of the river (T0a, T0t/k, and T1) compared to the T3 site which was located near the western bank. In general, the annual and diurnal cycles during 2014 were representative of the 2000-2013 distributions. The only exceptions were in March when the monthly mean rainrate was above the 95th percentile and September when both rain frequency and intensity were suppressed. The natural spatial variability must be accounted for before interpreting anthropogenically-induced differences among the GoAmazon2014/5 sites.

Amazon↗

Charging infrastructure access and operation to reduce the grid impacts of deep electric vehicle adoption

Electric vehicles will contribute to emissions reductions in the United States, but their charging may challenge electricity grid operations. We present a data-driven, realistic model of charging demand that captures the diverse charging behaviours of future adopters in the US Western Interconnection. We study charging control and infrastructure build-out as critical factors shaping charging load and evaluate grid impact under rapid electric vehicle adoption with a detailed economic dispatch model of 2035 generation. We find that peak net electricity demand increases by up to 25% with forecast adoption and by 50% in a stress test with full electrification. Locally optimized controls and high home charging can strain the grid. Shifting instead to uncontrolled, daytime charging can reduce storage requirements, excess non-fossil fuel generation, ramping and emissions. Our results urge policymakers to reflect generation-level impacts in utility rates and deploy charging infrastructure that promotes a shift from home to daytime charging.

33 ADVANCED PROPULSION SYSTEMS↗

PowerNet: Multi-agent Deep Reinforcement Learning for Scalable Powergrid Control

This paper develops an efficient multi-agent deep reinforcement learning algorithm for cooperative controls in powergrids. Specifically, we consider the decentralized inverter-based secondary voltage control problem in distributed generators (DGs), which is first formulated as a cooperative multi-agent reinforcement learning (MARL) problem. We then propose a novel on-policy MARL algorithm, PowerNet, in which each agent (DG) learns a control policy based on (sub-)global reward but local states and encoded communication messages from its neighbors. Motivated by the fact that a local control from one agent has limited impact on agents distant from it, we exploit a novel spatial discount factor to reduce the effect from remote agents, to expedite the training process and improve scalability. Furthermore, a differentiable, learning-based communication protocol is employed to foster the collaborations among neighboring agents. In addition, to mitigate the effects of system uncertainty and random noise introduced during on-policy learning, we utilize an action smoothing factor to stabilize the policy execution. To facilitate training and evaluation, we develop PGSim, an efficient, high-fidelity powergrid simulation platform. Here, experimental results in two microgrid setups show that the developed PowerNet outperforms the conventional model-based control method, as well as several state-of-the-art MARL algorithms. The decentralized learning scheme and high sample efficiency also make it viable to large-scale power grids.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Naturalness and the muon magnetic moment

We study a predictive model for explaining the apparent deviation of the muon anomalous magnetic moment from the Standard Model expectation. There are no new scalars and hence no new hierarchy puzzles beyond those associated with the Higgs; the only new particles at the TeV scale are vector-like singlet and doublet leptons. Interestingly, this simple model provides a calculable example violating the Wilsonian notion of naturalness: despite the absence of any symmetries prohibiting its generation, the coefficient of the naively leading dimension-six operator for (g – 2) vanishes at one-loop. While effective field theorists interpret this either as a surprising UV cancellation of power divergences, or as a delicate cancellation between matching UV and calculable IR corrections to (g – 2) from parametrically separated scales, there is a simple explanation in the full theory: the loop integrand is a total derivative of a function vanishing in both the deep UV and IR. The leading contribution to (g – 2) arises from dimension-eight operators, and thus the required masses of new fermions are lower than naively expected, with a sizeable portion of parameter space already covered by direct searches at the LHC. The viable parameter space free of fine-tuning for the muon mass will be fully covered by future direct LHC searches, and all of the parameter space can be probed by precision measurements at planned future lepton colliders.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

De novo Nanopore read quality improvement using deep learning

BACKGROUND: Long read sequencing technologies such as Oxford Nanopore can greatly decrease the complexity of de novo genome assembly and large structural variation identification. Currently Nanopore reads have high error rates, and the errors often cluster into low-quality segments within the reads. The limited sensitivity of existing read-based error correction methods can cause large-scale mis-assemblies in the assembled genomes, motivating further innovation in this area. RESULTS: Here we developed a Convolutional Neural Network (CNN) based method, called MiniScrub, for identification and subsequent "scrubbing" (removal) of low-quality Nanopore read segments to minimize their interference in downstream assembly process. MiniScrub first generates read-to-read overlaps via MiniMap2, then encodes the overlaps into images, and finally builds CNN models to predict low-quality segments. Applying MiniScrub to real world control datasets under several different parameters, we show that it robustly improves read quality, and improves read error correction in the metagenome setting. Compared to raw reads, de novo genome assembly with scrubbed reads produces many fewer mis-assemblies and large indel errors. CONCLUSIONS: MiniScrub is able to robustly improve read quality of Oxford Nanopore reads, especially in the metagenome setting, making it useful for downstream applications such as de novo assembly. We propose MiniScrub as a tool for preprocessing Nanopore reads for downstream analyses. MiniScrub is open-source software and is available at https://bitbucket.org/berkeleylab/jgi-miniscrub .

59 BASIC BIOLOGICAL SCIENCES↗

Search for vector-like leptons coupling to first- and second-generation Standard Model leptons in pp collisions at s = 13 TeV with the ATLAS detector

A search for pair production of vector-like leptons coupling to first- and second-generation Standard Model leptons is presented. The search is based on a dataset of proton-proton collisions at s$$ \sqrt{s} $$ = 13 TeV recorded with the ATLAS detector during Run 2 of the Large Hadron Collider, corresponding to an integrated luminosity of 140 fb−1. Events are categorised depending on the flavour and multiplicity of leptons (electrons or muons), as well as on the scores of a deep neural network targeting particular signal topologies according to the decay modes of the vector-like leptons. In each of the signal regions, the scalar sum of the transverse momentum of the leptons and the missing transverse momentum is analysed. The main background processes are estimated using dedicated control regions in a simultaneous fit with the signal regions to data. No significant excess above the Standard Model background expectation is observed and limits are set at 95% confidence level on the production cross-sections of vector-like electrons and muons as a function of the vector-like lepton mass, separately for SU(2) doublet and singlet scenarios. The resulting mass lower limits are 1220 GeV (1270 GeV) and 320 GeV (400 GeV) for vector-like electrons (muons) in the doublet and singlet scenarios, respectively.

Aad, G↗