Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “machine vision”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

AI foundation models for experimental fusion tasks

Artificial Intelligence (AI) foundation models, while successful in various domains of language, speech, and vision, have not been adopted in production for fusion energy experiments. This brief paper presents how AI foundation models can be used for fusion energy diagnostics, enabling, for example, visual automated logbooks to provide greater insights into chains of plasma events in a discharge, in time for between-shot analysis.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Data associated with “Different methods of estimating riverbed sediment grain size diverge at the basin scale ” (v2)

This data package is associated with the publication “Different methods of estimating riverbed sediment grain size diverge at the basin scale” published in Frontiers in Earth Science (Regier et al., 2025). The distribution of sediment grain size in streams and rivers is often quantified by the median grain size (d50), a key metric for understanding and predicting hydrologic and biogeochemical function of streams and rivers. Manual methods to measure d50 are time-consuming and ignore larger grains, while model-based methods to estimate d50 often over-generalize basin characteristics, and therefore cannot accurately represent site-scale heterogeneity. Here, we apply a machine learning-enabled photogrammetry methodology (You Only Look Once, or YOLO) for estimating d50 for grains > 2 mm based on images collected from streams and rivers throughout the Yakima River Basin (YRB). To understand how such methods may help bridge the gaps in resolution and accuracy between manual and catchment characteristics model-based d50 estimates, we compared YOLO d50 values to manual and model-based estimates across the YRB. We found distinct differences among methods for d50 averages and variability, and relationships between d50 estimates and basin characteristics. Source images can be found at https://data.ess-dive.lbl.gov/view/doi:10.15485/1892052. This data package was originally published in May 2023. It was updated August 2025 (v2; new and modified files). File and folder names were not revised to indicate changes. See the change history section in the readme for more details. In addition to the readme, this data package also includes a file-level metadata (FLMD) file that describes each file and a data dictionary (DD) that describes all column/row headers and variable definitions. This dataset is comprised of one main data folder containing (1) file-level metadata; (2) data dictionary; (3) readme; (4) and subfolders containing data, figures, and scripts. The data folder contains datasets used for the analyses in the manuscript in image, text-delimited or geospatially-referenced formats. The figures folder contains the figures from the manuscript in different formats. The scripts folder contains all of the scripts used to complete the analyses in the manuscript. All files are .csv, .rds, .dbf, .prj, .shp, .shx, .jpg, .png, .R, .Rproj, or .pdf. We acknowledge the Yakama Nation as owners and caretakers of the lands where we collected some of these data. We thank the Confederated Tribes and Bands of the Yakama Nation Tribal Council and Yakama Nation Fisheries for working with us to facilitate sample collection and optimization of data usage according to their values and worldview.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

An overview of visualization and visual analytics applications in water resources management

Recent advances in information, communication, and environmental monitoring technologies have increased the availability, spatiotemporal resolution, and quality of water-related data, thereby leading to the emergence of many innovative big data applications. Among these applications, visualization and visual analytics, also known as the visual computing techniques, empower the synergy of computational methods (e.g., machine learning and statistical models) with human reasoning to improve the understanding and solution toward complex science and engineering problems. These approaches are frequently integrated with geographic information systems and cyberinfrastructure to provide new opportunities and methods for enhancing water resources management. Here, we present a comprehensive review of recent hydroinformatics applications that employ visual computing techniques to (1) support complex data-driven research problems, and (2) support the communication and decision-makings in the water resources management sector. Then, we conduct a technical review of the state-of-the-art web-based visualization technologies and libraries to share our experiences on developing shareable, adaptive, and interactive visualizations and visual interfaces for water resources management applications. We close with a vision that applies the emerging visual computing technologies and paradigms to develop the next generation of hydroinformatics applications.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Demonstrating Advanced Sensors for In-Situ Monitoring Towards Qualification of Nuclear Relevant Components

The U.S. Department of Energy’s Office of Nuclear Energy Advanced Materials and Manufacturing Technologies (AMMT) program is pursuing qualification of laser powder bed fusion (LPBF) components for nuclear applications. A major focus of this effort is the use of in situ process monitoring and machine learning–based tools to establish real-time quality assurance. The primary objective of this report is to identify and evaluate the most relevant in situ sensor systems for LPBF, and to document the deployment of these systems across platforms critical to the AMMT program. This work demonstrates how in situ monitoring can detect process anomalies, track geometry-dependent flaws, and identify limiting combinations of processing parameters—particularly those related to energy density and complex geometries (e.g., overhanging structures). To support this goal, a diverse suite of sensor modalities was evaluated across LPBF platforms, including visible and near-infrared (NIR) imaging, fringe projection profilometry, long-wavelength infrared (LWIR) thermography, and high-speed photodiode/pyrometry systems. These sensor streams were integrated with Peregrine, a machine-agnostic software platform that, among other capabilities, can generate real-time process anomaly classification. This report documents sensor deployments on multiple AMMT flagship platforms, including the Concept Laser M2 and Renishaw AM400/AM250 systems. Calibration builds with complex, flaw-prone geometries such as unsupported overhangs, stepped features, and thin walls, were used to evaluate how well Peregrine and its associated sensors could detect process anomalies and other instabilities under varied energy densities. It will be shown how Peregrine reliably identifies common process anomalies such as recoater streaking, superelevation, etc., and can be used in post-build analysis for anomaly spatial distributions throughout the build height to better understand the impact of geometry and processing parameter choice on the build. This work demonstrates measurable progress toward the vision that components can be born-qualified by establishing a real-time monitoring framework, identifying limiting process conditions, and laying the foundation for sensor fusion–enabled prediction pipelines that are scalable across platforms and applicable to nuclear-relevant components.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

FY25 Electric Grid Security Annual Report

Sandia’s Electric Grid Security program advances a national vision of energy dominance and accessibility, while applying our national security -emphasis on ensuring of a secure, resilient, and affordable electric system for all users. Our achievements reflect a strategic approach combining technology development; modeling, simulation, and data analytics; and partnered demonstrations and outreach to further the adoption of advanced grid and storage technologies. Our FY25 efforts leverage the strengths of our partnerships—spanning Sandia’s core science and technology competencies as well as external technology leaders—to develop the solutions today which enable the grid of tomorrow. Key accomplishments in this report that support our strategy span our technical program areas and include: • New open-source analytical tools for systems -level planning and optimization, including significant advances to the QuESt analytical environment; • Further advancement of artificial intelligence and machine learning to enhanced grid operations and planning as we rise to the challenge of new large loads; • Development of solid-state power conversion technologies and a new medium-voltage research lab; • New technologies to assess wildfire vulnerabilities and mitigate potential impacts; • Advanced applications of new cybersecurity technologies with industry partners; • Contributions to understanding the impacts of electromagnetic pulses and geomagnetic disturbances on grid components; and • Digital twin development for hybrid microgrids with multiple generators, storage, and loads. This report indicates key areas of research and engagement and summarizes the impact of Sandia’s contributions through notable accomplishments, journal publications, patents, and technical conferences and presentations. It is provided with the hope that readers discover ways we can further team to create our modern grid and apply the outcomes of our efforts. The bulk of work described herein is funded by several offices within the U.S. Department of Energy (USDOE), including the Office of Electricity (OE); Cybersecurity, Energy Security, and Emergency Response (CESER); former offices such as the Office of Energy Efficiency and Renewable Energy (EERE), the Grid Deployment Office (GDO), the Office of Clean Energy Demonstrations (OCED), and other key programs at USDOE. As we continue to state in these annual reports, the contributors to our successes are too numerous to name here, though our team wishes to express our deep gratitude to the numerous program and project sponsors at the US Department of Energy, who often function equally as technical collaborators; our many partners in industry, academia, utilities, and other national labs; and fellow researchers and business partners at Sandia whose leadership and creativity have enabled the accomplishments described herein.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Convolutional neural networks for intra-hour solar forecasting based on sky image sequences

Accurate and timely solar forecasts play an increasingly critical role in power systems. Compared to longer forecasting timescales, very short-term solar forecasting has lagged behind in both research and practice. In this paper, we propose deep convolutional neural networks (CNNs) to provide operational intra-hour (10-minute-ahead to 60-minute-ahead) solar forecasts. We develop two CNN structures inspired by a widely-used CNN architecture. The CNNs are tailored to our solar forecasting regression tasks and rely solely on sky image sequences. Case studies based on six years of data (over 150,000 data points) demonstrate that the best CNN model has forecast skill scores of 20%-39% over the naive persistence of cloudiness benchmark, even at these very short timescales. The CNNs also have consistently superior performance when compared to shallow machine learning models with meteorological predictors, where the improvement averages around 7%. The sensitivity analyses show that the sky image length, resolution, and weather conditions have impacts on the deep learning model accuracy. In our intra-hour problem with specific setups, two sky images with a 10-minute 128 x 128 resolution yield the most accurate forecasts. Current limitations, future work, and deployment challenges and solutions are also discussed.

14 SOLAR ENERGY↗
DOE OSTI2023

Benchmarking for AI for Science

AI has been instrumental for recent developments in a number of domains of the sciences. With several hundred machine learning (ML) algorithms and models, and numerous AI-specific hardware platforms, a common quest for all scientists working on AI for Science is around the selection of machine learning algorithm(s) to solve their domain-specific scientific problems. A number of different initiatives around AI Benchmarking have been set up and have been useful in understanding the benefits of different ML algorithms for different tasks.However, with the majority of these AI Benchmarking initiatives focusing on the conventional notions of benchmarking, where the focus is purely runtime performance (such as training time or inference time), their suitability for benchmarking different ML algorithms for solving scientific problems has been viewed as a performance problem even though both are hardly the same. To make reasonable, explainable, and justifiable advancements in science using AI, it is critical to focus on the merits of these algorithms in handling different domain science problems. In other words, more emphasis must be given on Benchmarking for AI for Science than AI Benchmarking. The vision of the former is not only to assess the performance of ML algorithms, but also to assess, and understand the benefits and merits of different ML algorithms in handling scientific problems. Benchmarking for AI for Science, instead of pure performance focused AI Benchmarking, has several benefits: (i) it has the potential to offer advances in the sciences, much more rapidly than through pure performance-based AI methods, (ii) it will encourage the community to focus on developing better domain-specific AI techniques, particularly given the provision for being able to benchmark different techniques, and (iii) it will encourage hardware manufacturers to focus on developing science-specific hardware subsystems.

Thiyagalingam, Jeyan↗
DOE OSTI2023

MyCrunchGPT: A LLM Assisted Framework for Scientific Machine Learning

Scientific machine learning (SciML) has advanced recently across many different areas in computational science and engineering. Here, the objective is to integrate data and physics seamlessly without the need of employing elaborate and computationally taxing data assimilation schemes. However, preprocessing, problem formulation, code generation, postprocessing, and analysis are still time- consuming and may prevent SciML from wide applicability in industrial applications and in digital twin frameworks. Here, we integrate the various stages of SciML under the umbrella of ChatGPT, to formulate MyCrunchGPT, which plays the role of a conductor orchestrating the entire workflow of SciML based on simple prompts by the user. Specifically, we present two examples that demonstrate the potential use of MyCrunchGPT in optimizing airfoils in aerodynamics, and in obtaining flow fields in various geometries in interactive mode, with emphasis on the validation stage. To demonstrate the flow of the MyCrunchGPT, and create an infrastructure that can facilitate a broader vision, we built a web app based guided user interface, that includes options for a comprehensive summary report. The overall objective is to extend MyCrunchGPT to handle diverse problems in computational mechanics, design, optimization and controls, and general scientific computing tasks involved in SciML, hence using it as a research assistant tool but also as an educational tool. While here the examples focus on fluid mechanics, future versions will target solid mechanics and materials science, geophysics, systems biology, and bioinformatics.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

In-Situ Monitoring Assisted Large-Scale Additive Manufacturing of Mild Steel and 316L Alloys for Nuclear Application

The Advanced Materials and Manufacturing Technologies (AMMT) Program is aimed at developing cross-cutting technologies in support of a broad range of nuclear reactor parts, and to maintain U.S. leadership in materials and manufacturing technologies for nuclear energy applications. The overarching vision of the AMMT program is to accelerate the development, qualification, demonstration and deployment of advanced materials and manufacturing technologies to enable reliable and economical nuclear energy. One of its three goals is to target big challenges and game-changing technologies, to realize the mission and vision of AMMT program. Based on this context, this multi-year work package focuses on understanding the current state of large-scale additive manufacturing (AM) technology for the deposition of 316L stainless steel materials for final components and mild steel for use in nuclear manufacturing processes. The targeted AM modality is directed energy deposition (DED), capable of fabricating components on the size scale of meters including valves, pumps, impellers. etc. that are challenging or difficult to source, especially when developing new systems or replacing obsolete components. Accordingly, the current writeup aims at providing a baseline literature survey on structure-property relationships in mild steel and 316L alloys. Also, information on preliminary trials to date involving these two alloys show tremendous potential of printing parts having complex geometry and thin- walled structures, such as nuclear valve and Hot isostatic Press (HIP) can, using wire based (Wire Arc Additive Manufacturing and Hybrid Additive Manufacturing) as well as blown powder DED machines. All of this is aimed towards (i) demonstrating the ability to fabricate large components for pressure boundary applications relevant to the nuclear community and nuclear manufacturing technology, and (ii) understanding the effect of different manufacturing technology on AM can production and post HIPed material for nuclear applications. Another target of this writeup is to compile various in-situ monitoring tools that have been incorporated for different DED AM modalities, in order to understand process variability during the entire fabrication process. This can be correlated with processing-structure-property response surfaces and would add confidence around process quality verification and ultimately component certification for nuclear applications.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2023

Capturing dynamical correlations using implicit neural representations

Abstract Understanding the nature and origin of collective excitations in materials is of fundamental importance for unraveling the underlying physics of a many-body system. Excitation spectra are usually obtained by measuring the dynamical structure factor, S ( Q , ω ), using inelastic neutron or x-ray scattering techniques and are analyzed by comparing the experimental results against calculated predictions. We introduce a data-driven analysis tool which leverages ‘neural implicit representations’ that are specifically tailored for handling spectrographic measurements and are able to efficiently obtain unknown parameters from experimental data via automatic differentiation. In this work, we employ linear spin wave theory simulations to train a machine learning platform, enabling precise exchange parameter extraction from inelastic neutron scattering data on the square-lattice spin-1 antiferromagnet La 2 NiO 4 , showcasing a viable pathway towards automatic refinement of advanced models for ordered magnetic systems.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Automatic identification and quantification of dense microcracks in high-performance fiber-reinforced cementitious composites through deep learning-based computer vision

Highlights: • A method is presented to detect, locate, quantify, and visualize dense microcracks in HPFRCC. • Quantification of dense microcracks is realized using deep learning method for the first time. • The presented method uses deep learning models that are trained using a realistic dataset size. • The presented method has a high computation efficiency for identifying and quantifying cracks. • The presented method provides crack width with errors up to 50 μm and a R{sup 2} value of 0.984. High-performance fiber-reinforced cementitious composites (HPFRCCs) feature high mechanical strengths, crack resistance, and durability. Under excessive loading, HPFRCCs demonstrate dense microcracks that are difficult to identify using existing methods. This study presents a computer vision method for identification, quantification, and visualization of microcracks in HPFRCCs based on deep learning. The presented method integrates multiple deep learning models and computer vision techniques in a hierarchical architecture. The crack pattern (e.g., number, width, and spacing of cracks) are automatically determined from pictures without human intervention. This study shows that the presented method achieves an accuracy of 0.992 for crack detection and an accuracy finer than 50 μm (R{sup 2} > 0.984) for quantification of crack width when deep learning models are trained using only 200 pictures of HPFRCCs and 200 pictures of conventional concrete with incorporation of data augmentation. The presented method is expected to be also applicable to other materials featuring complex cracks.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Towards a Robust Adaptive Digital Twin for Fusion Applications

The development of a digital twin system for fusion applications is essential for enhancing the prediction, analysis, and optimization of complex plasma processes. Machine learning (ML), particularly deep learning has demonstrated strong capabilities in modeling such highly nonlinear and intricate systems. However, two critical challenges limit the deployment of deep learning-based digital twins: Uncertainty Quantification (UQ) and data drift. UQ is vital for ensuring trustworthy predictions, especially in decision-support scenarios. Additionally, data-driven models are often sensitive to changes in the underlying data distribution, such as shot-to-shot variations in fusion experiments, which can lead to performance degradation over time. To address these challenges, we are developing an uncertainty-aware, adaptive digital twin framework. Our approach incorporates deep learning models enhanced with Gaussian Process approximations for predictive uncertainty estimation, coupled with an online learning mechanism that enables continuous model adaptation to new experimental data. This adaptive capability allows the data driven models to respond effectively to evolving plasma behaviors and equipment conditions. Specifically, to mitigate the effects of shot-to-shot drift, our system updates itself incrementally as new data becomes available, improving both robustness and fidelity. Our vision is to evolve this data driven model into a self-sustaining digital twin system that leverages UQ based feedback to continuously refine itself and potentially support real-time decision making. This presentation will cover a brief background on uncertainty quantification for ML, our ongoing effort on development of UQ capabilities for ML, our data science pipeline from data collection to model development and analysis and online learning framework for modeling coil deflection at DIII-D. I will also briefly touch upon opportunities and challenges in development of digital twin framework.

Sammuli, Brian [General Atomics]↗
DOE OSTI2024

EFRC Accomplishments: Integrated Mesoscale Architectures for Sustainable Catalysis (IMASC)

The vision of IMASC is to advance the fundamental science necessary to alter fundamental aspects of the chemical industries sector and its carbon footprint. The IMASC mission is to develop design principles for increasing the efficiency of key chemical syntheses using dilute alloy catalysts by quantitatively scaling from model studies to catalytic operation using advanced experiment and theory. One of the major challenges in catalysis is predicting catalytic behavior—especially selectivity. The IMASC team developed a methodology for obtaining critical information for selected catalytic processes using dilute alloy catalysts, that is widely applicable to modeling catalytic processes.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Integrated Research Infrastructure Architecture Blueprint Activity (Final Report 2023)

The complexity of scientific pursuits is increasing rapidly with aspects that require dynamic integration of experiment, observation, theory, modeling, simulation, visualization, machine learning (ML), artificial intelligence (AI), and analysis. Research projects across the Department of Energy (DOE) are increasingly data and compute intensive. Innovative research teams are accelerating the pace of discovery by using high-performance computational and data tools in their research workflows and leveraging multiple research infrastructures. Additionally, several recent high-level U.S. government reports underscore the necessity of a new advanced computing ecosystem for international competitiveness and national security. International competitors are moving forward with major research infrastructure integration efforts that seek to capture a competitive advantage in the global innovation race. Owing to its unparalleled constellation of world-class experimental and observational facilities and high-performance and extreme-scale computational, data, and networking infrastructure, DOE is positioned to be a global leader in this new era of integrated science. However, this new integration paradigm will demand continuing evolution to ensure the U.S. remains a global leader in research and innovation. The DOE Office of Science (SC) has seized on the strategic importance of integration and has adopted a vision for Integrated Research Infrastructure (IRI): To empower researchers to meld DOE’s world-class research tools, infrastructure, and user facilities seamlessly and securely in novel ways to radically accelerate discovery and innovation. To respond to the evolving computational requirements of research and the competitive international innovation landscape, experimental facilities could be connected with high performance computing resources for near real-time analysis, and resources should be provided for merging enormous and diverse data for AI/ML techniques and analysis.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Secure Collaborative Environment for Seamless Sharing of Scientific Knowledge

In a secure collaborative environment, tera-bytes of data generated from powerful scientific instruments are used to train secure machine learning (ML) models on exascale computing systems, which are then securely shared with internal or external collaborators as cloud-based services. Devising such a secure platform is necessary for seamless scientific knowledge sharing without compromising individual, or institute-level, intellectual property and privacy details. By enabling new computing opportunities with sensitive data, we envision a secure collaborative environment that will play a significant role in accelerating scientific discovery. Several recent technological advancements have made it possible to realize these capabilities. In this paper, we present our efforts at ORNL toward developing a secure computation platform. We present a use case where scientific data generated from complex instruments, like those at the Spallation Neutron Source (SNS), are used to train a differential privacy enabled deep learning (DL) network on Summit, which is then hosted as a secure multi-party computation (MPC) service on ORNL’s Compute and Data Environment for Science (CADES) cloud computing platform for third-party inference. In this feasibility study, we discuss the challenges involved, elaborate on leveraged technologies, analyze relevant performance results and present the future vision of our work to establish secure collaboration capabilities within and outside of ORNL.

Yoginath, Srikanth↗
DOE OSTI2026

Integrated System Planning: Emerging Software Requirements in the Power Industry

Power system planning software remains fragmented across organizational boundaries, with specialized tools for capacity expansion, production cost modeling, power flow, and dynamic analysis operating on incompatible data models and assumptions. This article argues that the fragmentation is not merely a technical problem but a predictable consequence of Conway's law: software architectures mirror the departmental structures within which they are developed. Regulatory milestones like Federal Energy Regulatory Commission (FERC) Order 888 formalized these divisions, but the roots trace back to the distinct engineering disciplines-mechanical, chemical, and electrical-that staffed generation and transmission planning departments in vertically integrated utilities. As the industry moves toward integrated system planning (ISP) that coordinates generation, transmission, and distribution investment decisions, the software ecosystem must evolve accordingly. We identify five categories of software requirements to enable this transition: coherent data inputs decoupled from individual applications, unified and extensible data schemas, modular component representations that support multiple abstraction levels, lifecycle management of planning datasets, and well-defined application programming interface (API) contracts that separate data exchange from algorithmic control. We examine how these requirements interact with three common workflow patterns-serial gate clearing, sequential multiapplication, and convergence oriented-and discuss the interface design principles each demands. We then outline a vision for platform-based planning architectures where specialized analytical services compose through standardized interfaces and where artificial intelligence (AI)/machine learning (ML) tools augment decision support within a disciplined software infrastructure. The practices proposed here offer a path from today's siloed tool collections toward collaborative planning ecosystems capable of handling the complexity of modern power system transformation.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Wave Detection and Tracking Within a Rotating Detonation Engine Through Object Detection

As the operational time window of experimental rotating detonation engines (RDEs) is expanded and the technology matures toward integration within gas turbines, monitoring techniques must evolve to offer computationally efficient and highly time-resolved diagnostics. In this study, computer vision object detection methodology that seeks to reduce data processing time and calculate wave velocity within drastically reduced time intervals as compared to traditional high-frame-rate RDE images analysis techniques is proposed. The adapted you-only-look-once object detection network is trained to detect individual detonation waves within single down-axis RDE images. The wave location and rotational direction detected within a frame are tracked through a series of high-speed images to calculate the frame-to-frame wave velocity with the time-step resolution of $\mathrm{20 μs}$ across a series of frames. The analysis of the annotation box size and image linearization effects is presented, demonstrating the lowest frame-to-frame velocity total uncertainty of $\mathrm{±3.8\%}$ and the highest classification speed of 9.5 frames per second using linearized images. Linearized images “unwrap” the RDE annulus pixel region to a reduced image size. Here, this new method offers great reductions in data processing times and unsteady detonation behavior insight at intervals more comparable to the timescales of detonation wave interactions via the application of machine learning to experimental RDE data.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2025

Benchmarking a Tunable Quantum Neural Network on Trapped-Ion and Superconducting Hardware

We implement a quantum generalization of a neural network on trapped-ion and IBM superconducting quantum computers to classify MNIST images, a common benchmark in computer vision. The network feedforward involves qubit rotations whose angles depend on the results of measurements in the previous layer. The network is trained via simulation, but inference is performed experimentally on quantum hardware. The classical-to-quantum correspondence is controlled by an interpolation parameter, $a$, which is zero in the classical limit. Increasing $a$ introduces quantum uncertainty into the measurements, which is shown to improve network performance at moderate values of the interpolation parameter. We then focus on particular images that fail to be classified by a classical neural network but are detected correctly in the quantum network. For such borderline cases, we observe strong deviations from the simulated behavior. We attribute this to physical noise, which causes the output to fluctuate between nearby minima of the classification energy landscape. Such strong sensitivity to physical noise is absent for clear images. We further benchmark physical noise by inserting additional single-qubit and two-qubit gate pairs into the neural network circuits. Our work provides a springboard toward more complex quantum neural networks on current devices: while the approach is rooted in standard classical machine learning, scaling up such networks may prove classically non-simulable and could offer a route to near-term quantum advantage.

FOS: Physical sciences↗