Unsupervised Neural Network Adaptation at the Edge: A Design Space Exploration
Explore the source record for details and available documents.
Computer science↗
SEARCH · Engineering Papers
Results for “network science”
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.
Deep learning of free boundary and Stefan problems
Free boundary problems appear naturally in numerous areas of mathematics, science and engineering. These problems present a great computational challenge because they necessitate numerical methods that can yield an accurate approximation of free boundaries and complex dynamic interfaces. In this work, we propose a multi-network model based on physics-informed neural networks to tackle a general class of forward and inverse free boundary problems called Stefan problems. Specifically, we approximate the unknown solution as well as any moving boundaries by two deep neural networks. Besides, we formulate a new type of inverse Stefan problems that aim to reconstruct the solution and free boundaries directly from sparse and noisy measurements. We demonstrate the effectiveness of our approach in a series of benchmarks spanning different types of Stefan problems, and illustrate how the proposed framework can accurately recover solutions of partial differential equations with moving boundaries and dynamic interfaces. All code and data accompanying this manuscript are publicly available at https://github.com/PredictiveIntelligenceLab/DeepStefan.
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
Integrating Analytical Solutions and U-Net Model for Predicting Groundwater Contaminant Plumes in Pump-and-Treat Systems
Pump-and-treat (P&T) is a common technique for groundwater remediation involving the extraction and treatment of contaminated water above ground. Optimizing the design and operation of the P&T well network is essential for maximizing the system’s effectiveness and efficiency. However, this optimization often necessitates many model evaluations, leading to computationally demanding tasks. This study introduces a novel approach that integrates analytical solutions for groundwater dynamics with the U-Net (Ronneberger et al., 2015) deep learning framework to predict groundwater contaminant plume migration under dynamic pumping conditions. By incorporating the Thiem equation (Thiem, 1906) into the input preprocessing, the U-Net model transforms sparse well data into a continuous spatial field that captures the hydraulic impacts of pumping activities. This integration enables the model to leverage both deep learning capabilities and classical physics-based groundwater theories, enhancing prediction accuracy and computational efficiency. These advancements can facilitate rapid, large-scale evaluations of P&T optimization simulations, allowing for timely and effective decision-making in well placement and system management. We demonstrate the model's robust performance across both simplified transient 2D models and a more complex 3D heterogeneous site model at the 200 West P&T facility at the Hanford Site. The U-Net-based model offers substantial computational advantages, reducing simulation times significantly compared to full physics-based models and providing a powerful tool for rapid site evaluation and P&T system optimization, such as evaluating alternative P&T well network designs. Our findings highlight the potential of advanced machine learning models to significantly enhance the efficiency and sustainability of groundwater remediation efforts, offering a novel application of U-Net architecture in environmental science.
Pump-and-treat↗
High Velocity, Low‐Voltage Collective In‐Plane Switching in (100) BaTiO 3 Thin Films
Abstract Ferroelectrics are being increasingly called upon for electronic devices in extreme environments. Device performance and energy efficiency is highly correlated to clock frequency, operational voltage, and resistive loss. To increase performance it is common to engineer ferroelectric domain structure with highly‐correlated electrical and elastic coupling that elicit fast and efficient collective switching. Designing domain structures with advantageous properties is difficult because the mechanisms involved in collective switching are poorly understood and difficult to investigate. Collective switching is a hierarchical process where the nano‐ and mesoscale responses control the macroscopic properties. Using chemical solution synthesis, epitaxially nearly‐relaxed (100) BaTiO 3 films are synthesized. Thermal strain induces a strongly‐correlated domain structure with alternating domains of polarization along the [010] and [001] in‐plane axes and 90° domain walls along the [011] or [01] directions. Simultaneous capacitance–voltage measurements and band‐excitation piezoresponse force microscopy revealed strong collective switching behavior. Using a deep convolutional autoencoder, hierarchical switching is automatically tracked and the switching pathway is identified. The collective switching velocities are calculated to be ≈500 cm s −1 at 5 V (7 kV cm −1 ), orders‐of‐magnitude faster than expected. These combinations of properties are promising for high‐speed tunable dielectrics and low‐voltage ferroelectric memories and logic.
36 MATERIALS SCIENCE↗
Elucidating molecular scale interactions underlying the freezing behavior of salt solutions in silica nanopores
Elucidating the influence of nanoscale confinement on the freezing behavior of salt solutions is of fundamental interest to environmental security and materials science and engineering. Specific structural information such as the coordination environment of ions in confined aqueous medium, effects on the extended hexagonal network of water molecules and their mutual non-bonding interactions in confinement are sparse in the literature along with the change in the dynamical characteristics of water in the presence of ions. To address these knowledge gaps, the current study is focused on investigating the influence of reduced dimensionality arising from nanoscale confinement on the structural evolution of salt solutions on freezing and the associated fluid–surface interactions. In this regard, operando wide-angle X-ray scattering (WAXS) measurements and classical molecular dynamics (MD) simulations are conducted with water and 0.5 M CaCl 2 , MgCl 2 and KCl solutions confined in 4 nm sized SBA-15 silica pores upon cooling from 300 K to 200 K. The freezing point of the salt solutions is depressed to 235 K which is about 10 K lower than that of confined water. The translational dynamics of confined salt solutions indicates shifts in the fragile-to-strong dynamical crossover to a lower temperature compared to pure water. The strong electrostatic attraction between the cations and the surrounding water molecules contributes to the freezing point depression in confined salt solutions. These insights unlock the molecular-scale basis and mechanisms underlying the freezing behavior of confined salt solutions.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Intrinsic quality factor extraction of multi-port cavity with arbitrary coupling
In this work, we derived S-parameter based expressions for the intrinsic quality factor of an arbitrary coupled multiport microwave cavity and a non-ideal test fixture. Practical accuracy limitations of the obtained expressions specifically for superconducting accelerator cavities were evaluated both analytically and with the simulation software. The resulting formulas can be used to extract the intrinsic quality factor of normal-conducting and superconducting cavities directly from the calibrated S-parameter measurements.
43 PARTICLE ACCELERATORS↗
Structure and related properties of amorphous magnesium aluminosilicates
Not Available
36 MATERIALS SCIENCE↗
Semantic Segmentation of Micrographs for Nuclear Fuel Analysis and Degradation Quantification
Not Available
11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
Graph Analytics on Jellyfish topology
Because large unstructured datasets is important for many science domains, distributed graph analytics is critical to many scientists. Unfortunately, obtaining scaling and performance for irregular communication is challenging because contemporary network interconnects are primarily designed to maximize bandwidths of fixed-neighborhoods large-message exchanges (e.g., stencils). Although there is no consensus on the “best” network topologies for irregular communication, unstructured graph-based interconnects can be more suitable. We analyze three popular graph workloads – clustering, pattern enumeration, and traversal — on comparable networks (in terms of resources and costs) constructed from Jellyfish Random Regular, Dragonfly and Fat tree topologies, varying the routing algorithms. Using packet-level simulations, we demonstrate up to 60% improvement in communication time with Jellyfish due to diversity of the short paths between arbitrary endpoints, which can reduce overall network stalls and congestion.
Graph Analytics, network topology, interconnect, H↗
Microbial processing of plant remains is co-limited by multiple nutrients in global grasslands
Microbial processing of aggregate-unprotected organic matter inputs is key for soil fertility, long-term ecosystem carbon and nutrient sequestration and sustainable agriculture. In this work, we investigated the effects of adding multiple nutrients (nitrogen, phosphorus and potassium plus nine essential macro- and micro-nutrients) on decomposition and biochemical transformation of standard plant materials buried in 21 grasslands from four continents. Addition of multiple nutrients weakly but consistently increased decomposition and biochemical transformation of plant remains during the peak-season, concurrent with changes in microbial exoenzymatic activity. Higher mean annual precipitation and lower mean annual temperature were the main climatic drivers of higher decomposition rates, while biochemical transformation of plant remains was negatively related to temperature of the wettest quarter. Nutrients enhanced decomposition most at cool, high rainfall sites, indicating that in a warmer and drier future fertilized grassland soils will have an even more limited potential for microbial processing of plant remains.
59 BASIC BIOLOGICAL SCIENCES↗
Controls on Interspecies Electron Transport and Size Limitation of Anaerobically Methane-Oxidizing Microbial Consortia
Anaerobic oxidation of methane is a globally important, microbially mediated process reducing the emission of methane, a potent greenhouse gas.
59 BASIC BIOLOGICAL SCIENCES↗
A Pseudoreversible Normalizing Flow for Stochastic Dynamical Systems with Various Initial Distributions
Here, we present a pseudoreversible normalizing flow method for efficiently generating samples of the state of a stochastic differential equation (SDE) with various initial distributions. The primary objective is to construct an accurate and efficient sampler that can be used as a surrogate model for computationally expensive numerical integration of SDEs, such as those employed in particle simulation. After training, the normalizing flow model can directly generate samples of the SDE’s final state without simulating trajectories. The existing normalizing flow model for SDEs depends on the initial distribution, meaning the model needs to be retrained when the initial distribution changes. The main novelty of our normalizing flow model is that it can learn the conditional distribution of the state, i.e., the distribution of the final state conditional on any initial state, such that the model only needs to be trained once and the trained model can be used to handle various initial distributions. This feature can provide a significant computational saving in studies of how the final state varies with the initial distribution. Additionally, we propose to use a pseudoreversible network architecture to define the normalizing flow model, which has sufficient expressive power and training efficiency for a variety of SDEs in science and engineering, e.g., in particle physics. We provide a rigorous convergence analysis of the pseudoreversible normalizing flow model to the target probability density function in the Kullback–Leibler divergence metric. Numerical experiments are provided to demonstrate the effectiveness of the proposed normalizing flow model.
97 MATHEMATICS AND COMPUTING↗
Tencoder: tensor-product encoder-decoder architecture for predicting solutions of PDEs with variable boundary data
It is widely hoped that artificial intelligence will boost data-driven surrogate models in science and engineering. However, fundamental spatial aspects of AI surrogate models remain under-studied. We investigate the ability of neural-network surrogate models to predict solutions to PDEs under variable boundary values. We do not wish to retrain the model when the boundary values change but to make them inputs to the model and infer the solution of the PDE under those boundary conditions. Such a capability is essential to making AI-based surrogate models practically useful. While simple feedforward networks are used for one-dimensional (1D) Poisson equation, an encoder-decoder architecture with a tensor-product layer is developed for the two-dimensional Poisson equation posed on a rectangular domain. We show that it is indeed possible to infer solutions to PDEs from variable boundary data using neural networks in this relatively simple setting, and point to future directions.
Kashi, Aditya↗
The Low-Yield Nuclear Monitoring (LYNM) Experimental Science Plan
The Low-Yield Nuclear Monitoring (LYNM) Program is a long-term NNSA research and development effort designed to improve the United States’ explosion monitoring capabilities, particularly with respect to low-yield and potentially evasive underground nuclear testing. The LYNM Program focuses on researching, discovering, and exploiting unique and useful signatures, from all available technologies and sensors (e.g., seismic, acoustic, electromagnetic, gases, and particulates (both stable and radioactive)). Four Department of Energy laboratories participate in LYNM and together are referred to as the ‘quad-lab’. The R&D program execution is performed under NNSA defined structures known as “ventures”. Four science ventures were established: 1) Explosion Source Functions; 2) Containment of Low-Yield Underground Tests; 3) Local Signatures; 4) Dynamic Monitoring Networks. To organize and execute the large field scale experiment a fifth venture was established: 5) Physics Experiment One (PE-1). As part of the LYNM Program, a series of experiments are planned at various scales and levels of venture involvement. These vary from those that involve a subset of labs and/or LYNM ventures (e.g., small experiments), to full quad-lab LYNM Program field-scale integrated experiments. Such experiments may involve chemical explosions with tracer materials or other means of simulating the expected signals from a nuclear explosion. The LYNM Program does not conduct actual nuclear explosions. Since 1992, the U.S. has observed a moratorium on underground nuclear explosions. This document is intended to provide the underlying scientific basis for the LYNM planned experimental work. Each specific LYNM experiment will develop a goals, objectives, and requirements (GOR) plan following the guidance in this document. The LYNM technical staff will define the numbers and types of experiments required over the course of the Program based on technical needs and within funding constraints. As with any scientific experiment series, the number and types of experiments may change based upon the experimental results obtained. An experiment that agrees with models/codes/software signature predictions may need fewer repetitions/variations, depending upon the level of statistical rigor desired, as compared to one in which the predictions and experimental data do not match. The large LYNM field-scale integrated experiments require the longest lead-time for planning, and these are discussed in more detail near the end of this document.
58 GEOSCIENCES↗
TNet: A Model-Constrained Tikhonov Network Approach for Inverse Problems
Deep learning (DL), in particular deep neural networks, by default is purely data-driven and in general does not require physics. This is the strength of DL but also one of its key limitations when applied to science and engineering problems in which underlying physical properties—such as stability, conservation, and positivity—and accuracy are required. DL methods in their original forms are often not capable of respecting the underlying mathematical models or achieving desired accuracy even in big-data regimes. On the other hand, many data-driven science and engineering problems, such as inverse problems, typically have limited experimental or observational data, and DL would overfit the data in this case. Leveraging information encoded in the underlying mathematical models, we argue, not only compensates for missing information in low data regimes but also provides opportunities to equip DL methods with the underlying physics, hence promoting better generalization. This paper develops a model-constrained DL approach and its variant TNet—a Tikhonov neural network—which are capable of learning not only information hidden in the training data but also in the underlying mathematical models to solve inverse problems governed by partial differential equations in low data regimes. We provide the constructions and some theoretical results for the proposed approaches for both linear and nonlinear inverse problems. Since TNet is designed to learn inverse solutions with Tikhonov regularization, it is interpretable: in fact it recovers Tikhonov solutions for linear cases while potentially approximating Tikhonov solutions for nonlinear inverse problems. We also prove that data randomization can enhance not only the smoothness of the networks but also their generalizations. Comprehensive numerical results confirm the theoretical findings and show that with even as little as 1 training data sample for one-dimensional (1D) deconvolution, 5 for an inverse 2D heat conductivity problem, 100 for inverse initial conditions for a time-dependent 2D Burgers’s equation, and 50 for inverse initial conditions for 2D Navier–Stokes equations, TNet solutions can be as accurate as Tikhonov solutions while being several orders of magnitude faster. Furthermore, this is possible owing to the model-constrained term, replications, and randomization.
97 MATHEMATICS AND COMPUTING↗
Probing the self-ionization of liquid water with ab initio deep potential molecular dynamics
The chemical equilibrium between self-ionized and molecular water dictates the acid–base chemistry in aqueous solutions, yet understanding the microscopic mechanisms of water self-ionization remains experimentally and computationally challenging. Herein, Density Functional Theory (DFT)–based deep neural network (DNN) potentials are combined with enhanced sampling techniques and a global acid–base collective variable to perform extensive atomistic simulations of water self-ionization for model systems of increasing size. The explicit inclusion of long-range electrostatic interactions in the DNN potential is found to be crucial to accurately reproduce the DFT free energy profile of solvated water ion pairs in small (64 and 128 H 2 O) cells. The reversible work to separate the hydroxide and hydronium to a distance S is found to converge for simulation cells containing more than 500 H 2 O, and a distance of ∼ 8 Å is the threshold beyond which the work to further separate the two ions becomes approximately zero. The slow convergence of the potential of mean force with system size is related to a restructuring of water and an increase of the local order around the water ions. Calculation of the dissociation equilibrium constant illustrates the key role of long-range electrostatics and entropic effects in the water autoionization process.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Classification of Cloud Particle Imagery from Aircraft Platforms Using Convolutional Neural Networks
Abstract A vast amount of ice crystal imagery exists from a variety of field campaign initiatives that can be utilized for cloud microphysical research. Here, nine convolutional neural networks are used to classify particles into nine regimes on over 10 million images from the Cloud Particle Imager probe, including liquid and frozen states and particles with evidence of riming. A transfer learning approach proves that the Visual Geometry Group (VGG-16) network best classifies imagery with respect to multiple performance metrics. Classification accuracies on a validation dataset reach 97% and surpass traditional automated classification. Furthermore, after initial model training and preprocessing, 10 000 images can be classified in approximately 35 s using 20 central processing unit cores and two graphics processing units, which reaches real-time classification capabilities. Statistical analysis of the classified images indicates that a large portion (57%) of the dataset is unusable, meaning the images are too blurry or represent indistinguishable small fragments. In addition, 19% of the dataset is classified as liquid drops. After removal of fragments, blurry images, and cloud drops, 38% of the remaining ice particles are largely intersecting the image border (≥10% cutoff) and therefore are considered unusable because of the inability to properly classify and dimensionalize. After this filtering, an unprecedented database of 1 560 364 images across all campaigns is available for parameter extraction and bulk statistics on specific particle types in a wide variety of storm systems, which can act to improve the current state of microphysical parameterizations.
54 ENVIRONMENTAL SCIENCES↗
Deep active learning for classifying cancer pathology reports
Abstract Background Automated text classification has many important applications in the clinical setting; however, obtaining labelled data for training machine learning and deep learning models is often difficult and expensive. Active learning techniques may mitigate this challenge by reducing the amount of labelled data required to effectively train a model. In this study, we analyze the effectiveness of 11 active learning algorithms on classifying subsite and histology from cancer pathology reports using a Convolutional Neural Network as the text classification model. Results We compare the performance of each active learning strategy using two differently sized datasets and two different classification tasks. Our results show that on all tasks and dataset sizes, all active learning strategies except diversity-sampling strategies outperformed random sampling, i.e., no active learning. On our large dataset (15K initial labelled samples, adding 15K additional labelled samples each iteration of active learning), there was no clear winner between the different active learning strategies. On our small dataset (1K initial labelled samples, adding 1K additional labelled samples each iteration of active learning), marginal and ratio uncertainty sampling performed better than all other active learning techniques. We found that compared to random sampling, active learning strongly helps performance on rare classes by focusing on underrepresented classes. Conclusions Active learning can save annotation cost by helping human annotators efficiently and intelligently select which samples to label. Our results show that a dataset constructed using effective active learning techniques requires less than half the amount of labelled data to achieve the same performance as a dataset constructed using random sampling.
59 BASIC BIOLOGICAL SCIENCES↗