Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep direct use”

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 307 records · Page 17
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Basic Research Needs in The Science of Scientific Software Development and Use: Investment in Software is Investment in Science

Increasingly powerful and affordable computing has revolutionized scientific and scholarly discovery across a broad range of fields. Computing relies on software, which has been rapidly growing in scope, diversity, and complexity. At the same time, the methods, processes, and tools used to produce and utilize this essential software are often ad hoc, and the study and improvement of them are often done without the benefit of direct funding or prioritization. Consequently, concerns are growing about the productivity of the developers and users of scientific software, its sustainability, and the trustworthiness of the results that it produces. Increased investment, especially in the characterization and improvement of how scientific software is developed and used, is important for sustaining and improving the impact of software as the scope and complexity of scientific efforts expand. Without this investment, we face the risk of diminishing returns on our software investments because the demands for increased functionality, usability, reliability, and more will not be sufficiently met. The US Department of Energy Office of Science (DOE/SC) is at the forefront of modern software-enabled scientific discovery across numerous areas of computational, experimental, and observational science, including major investments in national user facilities that support these activities. For many years, DOE/SC software investments have provided tremendous value to the scientific community. We want to continue and further improve the value of DOE/SC software efforts by using a scientific approach to understanding and improving how scientific software is developed and used. In December 2021, the DOE/SC Office of Advanced Scientific Computing Research (ASCR) convened a workshop on basic research needs for the Science of Scientific-Software Development and Use (SSSDU). Through keynote presentations, lightning talks, and breakout groups, which built on insights from 124 pre-workshop position papers, participants discussed the current practice of software development, maintenance, evolution, and use, and considered how the scientific method could be used to examine these practices and develop more evidence-based approaches to enhance the impact of software and computing on all areas of science. Workshop participants identified three priority research directions (PRDs) and three important crosscutting themes that center on the following overarching insight: Software has become an essential part of modern science, impacting discoveries, policy, and technological development. To maintain and improve confidence in science delivered via software, we must improve the processes and tools that help us create and use software, and this enhancement requires a deep understanding of the diverse array of teams and individuals doing the work.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Dendrite-resolved, full-melt-pool phase-field simulations to reveal non-steady-state effects and to test an approximate model

In this report we study the epitaxial, columnar growth of (multiply oriented) dendrites/cells for a spot melt in a polycrystalline Al–Cu substrate using two-dimensional, phase-field, direct numerical simulations (DNS) at the full-melt-pool scale. Our main objective is to compare the expensive DNS model to a much cheaper but approximate “line” model in which a single-crystal phase-field simulation is confined to a narrow rectangular geometry. To perform this comparison, we develop algorithms that automatically extract quantities of interest (QoIs) from both DNS and line models. These QoIs allow us to quantitatively assess the assumptions in the line model and help us analyze its discrepancy with the DNS model. We consider four sets of heat source parameters, mimicking welding and additive manufacturing conditions, that create a combination shallow and deep melt pools. Our largest DNS simulation used 16K x 14K grid points in space. Our main findings can be summarized as follows. Under AM conditions, the QoIs of line models are in excellent agreement with the full DNS results for both shallow and deep melt pools. Under welding conditions, the primary spacing of the DNS model is smaller than the prediction of line model. We identify a geometric crowding effect that accounts for the discrepancies between the DNS and line models. We propose two potential mechanisms that determine the response of the microstructure to geometric crowding.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Microfluidic, One-Batch Synthesis of Pd Nanocrystals on N-Doped Carbon in Surfactant-Free Deep Eutectic Solvents for Formic Acid Electrochemical Oxidation

One of the grand challenges that impedes practical applications of nanomaterials is the lack of robust manufacturing methods that are scalable, cheap, and environmentally friendly. In this work, we address this challenge by developing a microfluidic approach that produces surfactant-free Pd nanocrystals (NCs) uniformly loaded on N-doped porous carbon in a one-batch process. The deep eutectic solvent (DES) prepared from choline chloride and ethylene glycol was employed as a novel synthesis solvent, and its extended hydrogen networks and abundant ionic species effectively stabilize Pd facets and confine nanocrystal sizes without using surfactants. The microreactors provide faster heat exchange and more uniform mass transport, which in combination with DES produced Pd NCs with better-defined shape and predominately exposed Pd (100) facet. Furthermore, we describe that the N-doped functional groups in porous carbon direct dense and uniform heterogeneous growth of Pd NCs in a one-batch process, thereby eliminating a separate catalyst deposition step that is often involved in conventional synthesis. The Pd NCs in the one-batch-produced Pd/C catalysts exhibited a size distribution of similar to ~13 ± 3.5 nm and a high ESCA of 46.0 m 2 /g and delivered 362 mA/mg for formic acid electrochemical oxidation with improved stability, demonstrating the unique potentials of microfluidic reactors and DES for the controllable and scalable synthesis of electrocatalyst materials for practical applications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Local Resistance Measurement for Degradation of c-Si Heterojunction with Intrinsic Thin Layer (HIT) Solar Modules

Silicon heterojunction with intrinsic thin layer (or HIT) modules typically degrade at a rate of less than 1% annually in solar fields with dominant degradation in open-circuit voltage and some degradation in series resistance. However, detailed mechanisms can differ from module to module. Here, we study increases in local series resistance that occur over long-term field deployment, indicated by cell areas where the photoluminescence intensity does not degrade but the electroluminescence degrades significantly. To directly measure the local series resistance, we have cored out the local electroluminescence-degraded area, and we measured the sheet resistance by 4-point-probe and local nm-scale resistance using scanning spreading resistance microscopy (SSRM). The results by 4-point-probe show scattered sheet resistance that can be caused, for example, by nonuniform current paths through the transparent conductive oxide layer, the a-Si:H emitter, or the near-junction c-Si inversion layer. In contrast, the SSRM results indicate a relatively uniform and non-degraded resistivity on smaller nanometer spatial scales. SSRM is an atomic force microscopy-based two-terminal resistance mapping technique that measures the local resistance in nm-volume beneath the probe. The consistent resistances measured on the control and degraded samples can exclude the degradation of transparent conductive oxide resistance.

41 EE - Solar Energy Technologies Office (EE-4S)↗
DOE OSTI2022

The Science of Scientific Software Development and Use

Increasingly powerful and affordable computing has revolutionized scientific and scholarly discovery across a broad range of fields. Computing relies on software, which has been rapidly growing in scope, diversity, and complexity. At the same time, the methods, processes, and tools used to produce and utilize this essential software are often ad hoc, and the study and improvement of them is often done without the benefit of direct funding or prioritization. Consequently, concerns are growing about the productivity of the developers and users of scientific software, its sustainability, and the trustworthiness of the results that it produces. The US Department of Energy Office of Science (DOE-SC) is at the forefront of modern software-enabled scientific discovery across numerous areas of computational, experimental, and observational science, including major investments in national user facilities that support these activities. In December 2021, the DOE-SC Office of Advanced Scientific Computing Research (ASCR) convened a workshop on basic research needs for the Science of Scientific Software Development and Use (SSSDU). Through keynote presentations, lightning talks, and breakout groups, participants discussed the current practice of software development, maintenance, evolution, and use, and considered how the scientific method could be used to examine these practices and develop more evidence-based approaches to enhance the impact of software and computing on all areas of science. Workshop participants identified three priority research directions (PRDs) and three important crosscutting themes that center on the following overarching insight: software has become an essential part of modern science that impacts new discovery, policy, and technological development. To have full confidence in science delivered via software, we must improve the processes and tools that help us create and use it, and this enhancement requires a deep understanding of the diverse array of teams and individuals doing the work. The full workshop report will be available at https://doi.org/10.2172/1846009.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Towards data-driven constitutive modelling for granular materials via micromechanics-informed deep learning

The analytical description of path-dependent elastic-plastic responses of a granular system is highly complicated because of continuously evolving microstructures and strain localisation within the system undergoing deformation. This study offers an alternative to the current analytical paradigm by developing micromechanics-informed machine-learning based constitutive modelling approaches for granular materials. A set of critical variables associated with the constitutive behaviour of granular materials are identified through an incremental stress-strain relationship analysis. Depending on the strategy to exploit the priori micromechanical knowledge, three different training strategies are explored. The first model uses only the measurable external variables to make stress predictions; the second model utilises a directed graph to link all the external strain sequences and internal microstructural evolution variables into a single prediction model comprised of a series of sub-mappings, and the third model explicitly integrates the physically important non-temporal properties with external strain paths into training through an enhanced Gated Recurrent Unit (GRU). These three models show satisfactory agreement with unseen test specimens based on multi-directional loading cases. The basic features and potential applications of each model are explained. Lastly, the key factors for constitutive training and limitations of the current work are also discussed in detail.

36 MATERIALS SCIENCE↗
DOE OSTI2024

A Survey on Privacy in Graph Neural Networks: Attacks, Preservation, and Applications

Graph Neural Networks (GNNs) have gained significant attention owing to their ability to handle graph-structured data and the improvement in practical applications. However, many of these models prioritize high utility performance, such as accuracy, with a lack of privacy consideration, which is a major concern in modern society where privacy attacks are rampant. To address this issue, researchers have started to develop privacy-preserving GNNs. Despite this progress, there is a lack of a comprehensive overview of the attacks and the techniques for preserving privacy in the graph domain. In this survey, we aim to address this gap by summarizing the attacks on graph data according to the targeted information, categorizing the privacy preservation techniques in GNNs, and reviewing the datasets and applications that could be used for analyzing/solving privacy issues in GNNs. We also outline potential directions for future research in order to build better privacy-preserving GNNs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Influence of carrier trapping on radiation detection properties in $\mathrm{CVD}$ grown 4H-SiC epitaxial layers with varying thickness up to 250 µm

We report low doped n-type 4H-SiC epitaxial layers of thickness 50, 150, and 250 µm grown by hot wall chemical vapor deposition were used to fabricate Ni/4H-SiC Schottky radiation detectors. The epitaxial layers were grown on the (0001) face of highly conductive bulk 4H-SiC substrates 8° offcut towards the $\langle 11\bar{2}1\rangle$ direction. The 50, 150, and 250 µm thick epilayer detectors, under optimized settings, showed energy resolutions of 2.0%, 0.78%, and 0.63%, respectively for 5.48 MeV alpha particles. Deep level transient spectroscopy studies showed that the observed variation in the detector resolution is linked to the defect parameters in the devices. Least-squares fitting of the bias dependence of the charge collection efficiency according to a drift-diffusion model, revealed minority carrier diffusion lengths of 16, 10, and 9.2 µm, respectively, implying that the detectors are not limited by minority carrier trapping. The detector performance was observed to be primarily dependent on the concentration and capture cross-sections of the lifetime killing electron traps Z 1/2 and EH 6/7 .

36 MATERIALS SCIENCE↗
DOE OSTI2020

Ionic conductive polymers as artificial solid electrolyte interphase films in Li metal batteries – A review

Lithium (Li) metal has been considered as the ultimate anode material for next-generation rechargeable batteries due to its ultra-high theoretical specific capacity (3860 mAh g -1 ) and the lowest reduction voltage (-3.04 V vs the standard hydrogen electrode). However, the dendritic Li formation, uncontrolled interfacial reactions, and huge volume variations lead to unstable solid electrolyte interphase (SEI) layer, low Coulombic efficiency and hence short cycling lifetime. Designing artificial solid electrolyte interphase (artificial SEI) films on the Li metal electrode exhibits great potential to solve the aforementioned problems and enable Li–metal batteries with prolonged lifetime. Polymer materials with good ionic conductivity, superior processability and high flexibility are considered as ideal artificial SEI film materials. In this review, according to the ionic conductive groups, recent advances in polymeric artificial SEI films are summarized to afford a deep understanding of Li ion plating/stripping behavior and present design principles of high-performance artificial SEI films in achieving stable Li metal electrodes. Perspectives regarding to the future research directions of polymeric artificial SEI films for Li–metal electrode are also discussed. The insights and design principles of polymeric artificial SEI films gained in the current review will be definitely useful in achieving the Li–metal batteries with improved energy density, high safety and long cycling lifetime toward next-generation energy storage devices.

25 ENERGY STORAGE↗
DOE OSTI2023

The Effect of DPC Fillers on FEPs Relevant to Disposal of SNF

The US Department of Energy (DOE) is investigating the use of different materials that could be used to fill the void space inside a dual-purpose canister (DPC) loaded with spent nuclear fuel (SNF) just before it is emplaced in a deep geologic repository. The purpose of adding filler material is to maintain subcritical conditions in the repository during the postclosure period, which can span up to 1,000,000 years. Several types of materials have been proposed, including metals, cements, particulates, and glass. Part of this investigation addresses how the presence of filler material inside a DPC will affect the performance of the repository with respect to the repository features; the consequences of events that may occur; and the multiple thermal, hydrologic, chemical, and mechanical processes that may occur in a deep geologic repository over long timescales. This report describes some of the filler materials that have been proposed and studied; identifies 11 features, 6 events, and 25 processes that may be affected by the presence of filler materials; and discusses the effects that may require consideration for each feature, event, or process. The results of this study can be used to direct appropriate research and to develop suitable models if the DOE decides to use fillers to maintain subcritical conditions in DPCs used to dispose of SNF.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2025

Triangle Method for Dense ReLU Layers [SWR-25-72]

This software is an implementation of the methods for initializing and training neural networks to be more efficient per parameter, described more fully below and in the related publication: In theory, depth should make a ReLU network EXPONENTIALLY more efficient by enabling it to produce an exponential number of piecewise linear sections in its output. This reasoning is largely based on the work of mathematicians that have hand-constructed networks that make good use of depth. In practice however, even very deep ReLU networks that have been randomly initialized will behave identically to their shallow counterparts - missing an entire exponential dimension of efficiency. The triangle method is a first attempt at realizing the exponential potential of deep networks. Instead of randomly setting weights, we force pairs of neurons in each layer learn to build triangles (i.e. functions from [0,1] -> [0,1] that look like triangles). This is a very efficient pattern for generating lots of linear pieces because composing two triangular functions doubles the number of pieces with each composition. The triangle method is more than just a different initialization, it is a new paradigm of training. Instead of making direct updates to the matrix weights, we do an extra step of backpropagation to collect the derivatives of the loss function with respect to the shapes of the triangles, training them to tilt left or right. This process essentially holds the networks hand throughout the loss landscape and forces it to always use depth effectively by producing triangular shapes internally. This can produce several orders of magnitude of improvement on convex one-dimensional regression problems. Much more theoretical work is needed to realize its full potential beyond this context, but the implementation in this repository will still work in arbitrary numbers of dimensions. The file Triangle_Method.py is a generalized form of the method that will build each neuron its own custom 1-d convex activation function (with exponential efficiency). Example usage on one dimensional problems can be found in Example_Usage.ipynb and an example of using this in a real neural network can be found in Example_VGG16_CIFAR10.ipynb.

Milkert, Max [National Renewable Energy Laboratory↗
DOE OSTI2023

Application of a deep learning semantic segmentation model to helium bubbles and voids in nuclear materials

Imaging nanoscale radiation-induced defects using the transmission electron microscope (TEM) is a key factor in the successful implementation of materials for nuclear energy structural applications. Analyzing each defect in a TEM micrograph is currently a manual task. To identify the defects in a single image can take anywhere from 15 min to an hour and a project can require the analysis of anywhere from tens to ≥ 100 images. Here, we use artificial intelligence (AI) models to automate this task. For simplification, we evaluated images with only a single type of defect; helium bubbles. Additionally, we performed semantic segmentation of these helium bubble defects in electron microscopy images of irradiated FeCrAl alloys using a deep learning DefectSegNet model. This model, which was previously used to classify crystal defects, is inspired by the classic DenseNet and U-Net image segmentation models. It claims high spatial resolution, but has poor performance at object boundaries. Our paper improves the DefectSegNet model’s application by adding two new features. First, the DefectSegNet model is applied not only to perform calculation pixel-wise but also object (or feature) wise. Because object-wise metrics are directly relevant to our final goal of detecting bubbles, whereas pixel-wise classification is only an intermediate step, it is an important part of our study. Second, a distance map loss (DML) function has been added to increase its performance at object boundaries. It is crucial to accurately represent defects boundaries, especially bubbles, in order to track the bubble-induced swelling caused by irradiation. The boundary-focused DML function is also compared to other loss functions like Cross-entropy, Weighted Binary Cross Entropy (WBCE), Dice and Intersection over Union (IOU). Finally, by incorporating new features, we found a marked improvement on segmentation quality and better shape preservation at the boundaries and areas of the bubbles.

42 ENGINEERING↗
DOE OSTI2021

Sculpting the Plasmonic Responses of Nanoparticles by Directed Electron Beam Irradiation

Spatial confinement of matter in functional nanostructures has propelled these systems to the forefront of nanoscience, both as a playground for exotic physics and quantum phenomena and in multiple applications including plasmonics, optoelectronics, and sensing. In parallel, the emergence of monochromated electron energy loss spectroscopy (EELS) has enabled exploration of local nanoplasmonic functionalities within single nanoparticles and the collective response of nanoparticle assemblies, providing deep insight into associated mechanisms. However, modern synthesis processes for plasmonic nanostructures are often limited in the types of accessible geometry, and materials and are limited to spatial precisions on the order of tens of nm, precluding the direct exploration of critical aspects of the structure-property relationships. Here, the atomic-sized probe of the scanning transmission electron microscope is used to perform precise sculpting and design nanoparticle configurations. Using low-loss EELS, dynamic analyses of the evolution of the plasmonic response are provided. It is shown that within self-assembled systems of nanoparticles, individual nanoparticles can be selectively removed, reshaped, or patterned with nanometer-level resolution, effectively modifying the plasmonic response in both space and energy. This process significantly increases the scope for design possibilities and presents opportunities for unique structure development, which are ultimately the key for nanophotonic design.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Unraveling Defect-Dependent Conductivity-Type Switching in CuFe 2 O 4 for Enhanced Photoelectrocatalytic Reduction of Benzaldehyde

Photoelectrocatalytic (PEC) reduction provides a sustainable route for upgrading biomass-derived feedstocks with reduced energy requirements, yet remains largely unexplored beyond hydrogen evolution and CO 2 reduction due to the scarcity of stable photocathodes. Here, we report a defect-engineered CuFe 2 O 4 photocathode that enables directly quantified PEC reduction of benzaldehyde to benzyl alcohol using a singlecomponent, earth-abundant oxide. By controlling annealing temperature and oxygen partial pressure, CuFe 2 O 4 is systematically tuned from n-type to p-type conductivity. Electrochemical measurements, X-ray and ultraviolet photoelectron spectroscopy, and firstprinciples defect calculations collectively show that oxygen-rich annealing conditions suppress deep donor-type oxygen vacancies while stabilizing shallow acceptor-type copper vacancies, resulting in enhanced hole concentration and improved charge transport. In a mixed acetonitrile/water electrolyte employing 1,4-benzoquinone as a redox mediator, the optimized CuFe 2 O 4 photocathode achieves stable photoelectrochemical operation over 18 h under continuous illumination with a benzyl alcohol production rate of 2.57 μmol/h at −0.50 V vs Ag/AgNO 3 , corresponding to a Faradaic efficiency of 51.3%. This PEC approach lowers the required applied potential by ∼1 V compared to traditional electrocatalytic methods, offering a more energy-efficient route for carbonyl reduction. These findings establish CuFe 2 O 4 as a viable photocathode platform for sustainable photoelectrocatalytic organic transformations under mild reaction conditions.

benzaldehyde reduction↗
DOE OSTI2024

Learning model combining convolutional deep neural network with a self-attention mechanism for AC optimal power flow

Alternating current optimal power flow (OPF) analysis is critical for efficient and reliable operation of power systems. For large systems or repetitive computations, the traditional methods such as the direct and gradient methods, or non-traditional methods, such as the genetic algorithm and simulating annealing, are time-consuming and unsuitable for real-time computing. The work in this paper proposes a novel framework to obtain the optimal solution of power flow in real-time using a combination of convolutional neural networks and a self-attention mechanism. All parameters of the power networks are rearranged in an image-like shape of a multi-channel image where each channel is a two-dimensional matrix. The proposed approach is adaptive with every input size of power systems as well as frequent variations of network topologies without intervention to the framework core. The encompassment of all power system contexts in which all parameters of internal elements, generation costs, and topology information are included, contributes to the higher accuracy of inference compared to other current machine-learning-based OPF-solving methods. Besides, the proposed framework established on ubiquitous platforms is effortlessly integrated into current infrastructures of power systems, and the great efficiency along with the computation speed may serve as a critical point for practical implications, such as enabling faster decision-making during real-time operations, predicting system contingencies, and remedial actions based on an offline pre-trained model. Furthermore, this supervised learning process is applied to the dataset of four case studies of meshed power systems: the IEEE 5-bus system (IEEE-5), the IEEE 30-bus system (IEEE-30), the IEEE 39-bus system (IEEE-39), and the IEEE 57-bus system (IEEE-57) to prove the efficacy of the proposed method.

42 ENGINEERING↗
DOE OSTI2023

Compute-Efficient Deep Learning: Algorithmic Trends and Opportunities

Although deep learning has made great progress in recent years, the exploding economic and environmental costs of training neural networks are becoming unsustainable. To address this problem, there has been a great deal of research on algorithmically-efficient deep learning, which seeks to reduce training costs not at the hardware or implementation level, but through changes in the semantics of the training program. In this paper, we present a structured and comprehensive overview of the research in this field. First, we formalize the algorithmic speedup problem, then we use fundamental building blocks of algorithmically efficient training to develop a taxonomy. Our taxonomy highlights commonalities of seemingly disparate methods and reveals current research gaps. Next, we present evaluation best practices to enable comprehensive, fair, and reliable comparisons of speedup techniques. To further aid research and applications, we discuss common bottlenecks in the training pipeline (illustrated via experiments) and offer taxonomic mitigation strategies for them. Finally, we highlight some unsolved research challenges and present promising future directions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Peripheral positions encode transport specificity in the small multidrug resistance exporters

In secondary active transporters, a relatively limited set of protein folds have evolved diverse solute transport functions. Because of the conformational changes inherent to transport, altering substrate specificity typically involves remodeling the entire structural landscape, limiting our understanding of how novel substrate specificities evolve. In the current work, we examine a structurally minimalist family of model transport proteins, the small multidrug resistance (SMR) transporters, to understand the molecular basis for the emergence of a novel substrate specificity. We engineer a selective SMR protein to promiscuously export quaternary ammonium antiseptics, similar to the activity of a clade of multidrug exporters in this family. Using combinatorial mutagenesis and deep sequencing, we identify the necessary and sufficient molecular determinants of this engineered activity. Using X-ray crystallography, solid-supported membrane electrophysiology, binding assays, and a proteoliposome-based quaternary ammonium antiseptic transport assay that we developed, we dissect the mechanistic contributions of these residues to substrate polyspecificity. We find that substrate preference changes not through modification of the residues that directly interact with the substrate but through mutations peripheral to the binding pocket. Our work provides molecular insight into substrate promiscuity among the SMRs and can be applied to understand multidrug export and the evolution of novel transport functions more generally.

Science & Technology - Other Topics↗
DOE OSTI2023

Integration of Ag-CBRAM crossbars and Mott ReLU neurons for efficient implementation of deep neural networks in hardware

In-memory computing with emerging non-volatile memory devices (eNVMs) has shown promising results in accelerating matrix-vector multiplications. However, activation function calculations are still being implemented with general processors or large and complex neuron peripheral circuits. Here, we present the integration of Ag-based conductive bridge random access memory (Ag-CBRAM) crossbar arrays with Mott rectified linear unit (ReLU) activation neurons for scalable, energy and area-efficient hardware (HW) implementation of deep neural networks. We develop Ag-CBRAM devices that can achieve a high ON/OFF ratio and multi-level programmability. Compact and energy-efficient Mott ReLU neuron devices implementing ReLU activation function are directly connected to the columns of Ag-CBRAM crossbars to compute the output from the weighted sum current. We implement convolution filters and activations for VGG-16 using our integrated HW and demonstrate the successful generation of feature maps for CIFAR-10 images in HW. Our approach paves a new way toward building a highly compact and energy-efficient eNVMs-based in-memory computing system.

Mott insulators↗