Search NASA⌕ Search

SEARCH · Search NASA

Results for “Network Models”

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 973 records · Page 54

Comparative Study of High-Resolution LysB29(Nε-myristoyl) des(B30) Insulin Structures Display Novel Dynamic Causal Interrelations in Monomeric-Dimeric Motions

The treatment of insulin-dependent diabetes mellitus is characterized by artificial supplementation of pancreatic β-cell ability to regulate sugar levels in the blood. Even though various insulin analogs are crucial for reasonable glycemic control, understanding the dynamic mechanism of the insulin analogs may help to improve the best-protracted insulin analog to assist people with type 1 diabetes (T1D) to live comfortably while maintaining tight glycemic control. Here, we present the high-resolution crystal structure of NN304, known as insulin detemir, to 1.7 Å resolution at cryogenic temperature. We computationally further investigated our crystal structure’s monomeric-dimeric conformation and dynamic profile by comparing it with a previously available detemir structure (PDB ID: 1XDA). Our structure (PDB ID: 8HGZ), obtained at elevated pH, provides electrostatically triggered minor movements in the equilibrium between alternate conformational substates compared to the previous structure, suggesting it might induce an intermediate state in the dissociation pathway of the insulin detemir’s hexamer:dihexamer equilibrium. Supplemented with orientational cross-correlation analysis by a Gaussian network model (GNM), this alternate oligomeric conformation offers the distinct cooperative motions originated by loose coupling of distant conformational substates of a protracted insulin analog that has not been previously observed.

59 BASIC BIOLOGICAL SCIENCES↗

Structured hints : extracting and abstracting domain expertise.

We propose a new framework for providing information to help optimize domain-specific application codes. Its design addresses problems that derive from the widening gap between the domain problem statement by domain experts and the architectural details of new and future high-end computing systems. The design is particularly well suited to program execution models that incorporate dynamic adaptive methodologies for live tuning of program performance and resource utilization. This new framework, which we call 'structured hints', couples a vocabulary of annotations to a suite of performance metrics. The immediate target is development of a process by which a domain expert describes characteristics of objects and methods in the application code that would not be readily apparent to the compiler; the domain expert provides further information about what quantities might provide the best indications of desirable effect; and the interactive preprocessor identifies potential opportunities for the domain expert to evaluate. Our development of these ideas is progressing in stages from case study, through manual implementation, to automatic or semi-automatic implementation. In this paper we discuss results from our case study, an examination of a large simulation of a neural network modeled after the neocortex.

97 MATHEMATICS AND COMPUTING↗

Live cell imaging of cellular dynamics in poplar wood using computational cannula microscopy

This study presents significant advancements in computational cannula microscopy for live imaging of cellular dynamics in poplar wood tissues. Leveraging machine-learning models such as pix2pix for image reconstruction, we achieved high-resolution imaging with a field of view of 55µm using a 50µm-core diameter probe. Our method allows for real-time image reconstruction at 0.29 s per frame with a mean absolute error of 0.07. We successfully captured cellular-level dynamics in vivo , demonstrating morphological changes at resolutions as small as 3µm. We implemented two types of probabilistic neural network models to quantify confidence levels in the reconstructed images. This approach facilitates context-aware, human-in-the-loop analysis, which is crucial for in vivo imaging where ground-truth data is unavailable. Using this approach we demonstrated deep in vivo computational imaging of living plant tissue with high confidence (disagreement score ⪅0.2). This work addresses the challenges of imaging live plant tissues, offering a practical and minimally invasive tool for plant biologists.

Ingold, Alexander (ORCID:0009000752380016)↗

Phobos' gravity field and its influence on its orbit and physical librations

A model describing the physical libration in longitude and latitude for Phobos is derived. The major effect is the well-known longitude variation with the anomalistic orbital period and amplitude. Several additional meter-sized periodic librations in longitude exist. The latitude variation is dominated by the forced precession of Phobos' figure axis with the precession of Phobos' orbital plane. The contribution of Phobos' topography to its gravity field is estimated using the control network model of Duxbury and Callahan (1989).

Borderies, N.↗

A reinforcement learning-based architecture for fuzzy logic control

This paper introduces a new method for learning to refine a rule-based fuzzy logic controller. A reinforcement learning technique is used in conjunction with a multilayer neural network model of a fuzzy controller. The approximate reasoning based intelligent control (ARIC) architecture proposed here learns by updating its prediction of the physical system's behavior and fine tunes a control knowledge base. Its theory is related to Sutton's temporal difference (TD) method. Because ARIC has the advantage of using the control knowledge of an experienced operator and fine tuning it through the process of learning, it learns faster than systems that train networks from scratch. The approach is applied to a cart-pole balancing system.

Berenji, Hamid R.↗

DDStore: Distributed Data Store for Scalable Training of Graph Neural Networks on Large Atomistic Modeling Datasets

Graph neural networks (GNNs) are a class of Deep Learning models used in designing atomistic materials for effective screening of large chemical spaces. To ensure robust prediction, GNN models must be trained on large volumes of atomistic data on leadership class supercomputers. Even with the advent of modern architectures that consist of multiple storage layers that include node-local NVMe devices in addition to device memory for caching large datasets, extreme-scale model training faces I/O challenges at scale.We present DDStore, an in-memory distributed data store designed for GNN training on large-scale graph data. DDStore provides a hierarchical, distributed, data caching technique that combines data chunking, replication, low-latency random access, and high throughput communication. DDStore achieves near-linear scaling for training a GNN model using up to 1000 GPUs on the Summit and Perlmutter supercomputers, and reaches up to a 6.15x reduction in GNN training time compared to state-of-the-art methodologies.

Choi, Jong Youl↗

Even Higher-Level Synthesis: An Exploration of AI Hardware Accelerators using HLS4ML

With the rise of artificial intelligence, the popularization of deep learning, and a constantly evolving industry, the demand for flexible and efficient tools has never been greater. As algorithms grow more complex, their runtime and energy consumption increase exponentially. Customized hardware accelerators, long used for specific mathematical operations, remain essential for managing modern applications' computational and power demands. Hardware accelerators can speed up complex computations by orders of magnitude, but their manual design and verification processes are often challenging and time-consuming. High-Level Synthesis (HLS) provides a solution by transforming high-level algorithm descriptions, typically written in C++ or SystemC, into synthesizable RTL suitable for hardware implementation. This approach reduces development time for RTL engineers while offering flexibility beyond what traditional handwritten RTL can provide. We extended this capability to the machine-learning domain with the open-source framework hls4ml, which allows neural networks trained in Python frameworks like Tensorflow or PyTorch to be synthesized into efficient hardware representations for the traditional FPGA and ASIC flows. This breakthrough addresses the growing need for reduced design turnaround and easy verification of ML hardware accelerators with low latency and power efficiency constraints. During this tutorial, we will demonstrate how Python complements HLS by simplifying the ML design process, bridging the gap between software and hardware development. Attendees will explore how we translate neural networks modeled in Python into fixed-point C++ models suitable for HLS workflows. We will dive into strategies like Value-Range Analysis and Quantization-Aware Training, which optimize these designs for deployment and evaluate their accuracy, power consumption, and energy efficiency. To exemplify these concepts, experts from Fermilab will share their experiences applying this technology to high-energy physics experiments, where real-time, low-latency processing is critical. Over the years, Fermilab engineers have demonstrated how deep neural networks, optimized for hardware using hls4ml, can meet the stringent requirements of trigger systems at the CERN Large Hadron Collider. These systems rely on rapid decision-making to process immense data volumes while retaining only the most relevant events for further analysis. The application of hls4ml has also been extended to innovative technologies like smart pixel arrays. These smart pixels integrate ML inference capabilities directly into sensor devices, enabling localized data processing at the pixel level. This approach drastically reduces the need to transmit raw data to external processing units, significantly decreasing power consumption and latency. By embedding neural networks within the pixel architecture, the smart pixels can identify and prioritize relevant data in real time, providing a highly efficient solution for edge computing in scenarios such as particle detectors and imaging systems. Fermilab's work highlights the potential of hardware-accelerated ML in scenarios where both speed and power efficiency are mission-critical. Through this tutorial, attendees will gain valuable insights into the challenges and solutions of deploying ML in hardware. Understanding how HLS and hls4ml streamline the development of neural network-based hardware accelerators is fundamental for the industry's future. Participants will learn how these technologies are shaping the future of AI and scientific computing.

Di Guglielmo, Giuseppe [Fermilab]↗

Short–Period Variables in TESS Full–Frame Image Light Curves Identified via Convolutional Neural Networks

The Transiting Exoplanet Survey Satellite (TESS) mission measured light from stars in ∼85% of the sky throughout its 2 yr primary mission, resulting in millions of TESS 30-minute-cadence light curves to analyze in the search for transiting exoplanets. To search this vast data set, we aim to provide an approach that is computationally efficient, produces accurate predictions, and minimizes the required human search effort. We present a convolutional neural network that we train to identify short-period variables. To make a prediction for a given light curve, our network requires no prior target parameters identified using other methods. Our network performs inference on a TESS 30-minute-cadence light curve in ∼5 ms on a single GPU, enabling large-scale archival searches. We present a collection of 14,156 short-period variables identified by our network. The majority of our identified variables fall into two prominent populations, one of close-orbit main-sequence binaries and another of δ Scuti stars. Our neural network model and related code are additionally provided as open-source code for public use and extension.

Convolutional neural networks↗

Parallel discrete event simulation: A shared memory approach

With traditional event list techniques, evaluating a detailed discrete event simulation model can often require hours or even days of computation time. Parallel simulation mimics the interacting servers and queues of a real system by assigning each simulated entity to a processor. By eliminating the event list and maintaining only sufficient synchronization to insure causality, parallel simulation can potentially provide speedups that are linear in the number of processors. A set of shared memory experiments is presented using the Chandy-Misra distributed simulation algorithm to simulate networks of queues. Parameters include queueing network topology and routing probabilities, number of processors, and assignment of network nodes to processors. These experiments show that Chandy-Misra distributed simulation is a questionable alternative to sequential simulation of most queueing network models.

Reed, Daniel A.↗

Implementation of a Binary Neural Network on a Passive Array of Magnetic Tunnel Junctions

The increasing scale of neural networks and their growing application space have produced demand for more energy- and memory-efficient artificial-intelligence-specific hardware. Avenues to mitigate the main issue, the von Neumann bottleneck, include in-memory and near-memory architectures, as well as algorithmic approaches. In this report we leverage the low-power and the inherently binary operation of magnetic tunnel junctions (MTJs) to demonstrate neural network hardware inference based on passive arrays of MTJs. In general, transferring a trained network model to hardware for inference is confronted by degradation in performance due to device-to-device variations, write errors, parasitic resistance, and nonidealities in the substrate. To quantify the effect of these hardware realities, we benchmark 300 unique weight matrix solutions of a two-layer perceptron to classify the Wine dataset for both classification accuracy and write fidelity. Despite device imperfections, we achieve software-equivalent accuracy of up to 95.3% with proper tuning of network parameters in 15 x 15 MTJ arrays having a range of device sizes. The success of this tuning process shows that new metrics are needed to characterize the performance and quality of networks reproduced in mixed signal hardware.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

OpenDSS-wrapper (Distribution System Co-simulator with Distributed Energy Resource Controls)

Electric grid transformation with the proliferation of distributed energy resources (DER), such as solar photovoltaic (PV), wind, advanced energy storage technologies, and electric vehicles, and the growing use of communication technologies in both transmission and distribution systems are increasing the need to capture the interactions among these systems. Advanced modeling, control, and simulation tools that can perform co-simulation of electric power systems with other domains become indispensable to accurately model these interactions. To address this need, we have developed a codebase that integrates an electric power distribution system simulator, DER models, and DER controls. The codebase is based on OpenDSS, a distribution system simulator, and an existing open source co-simulation framework called HELICS. The contribution and uniqueness of the proposed codebase is that it tailors the generic HELICS framework specifically for distribution grid-related applications. The codebase includes an OpenDSS wrapper that controls the simulation, implements advanced DER controls, and extracts power flows, voltages, and other power system element information from the the distribution network modeled in OpenDSS. Sample HELICS federates, including a federate for OpenDSS, are provided that can communicate messages through the HELICS interface. Sample federates can be modified and additional HELICS federates can be added by the user depending on their use case requirements. SEE ALSO: https://github.com/NREL/dss-cosim

Blonsky, Michael↗

Application of the Koopman Operator-Theoretic Framework to Power System Dynamic State Estimation

Model-based and data-driven methods are combined to develop a hierarchical decentralized, robust dynamic state estimator (DSE). A two-level hierarchy is proposed, where the lower level consists of robust, model-based, decentralized DSEs. The state estimates sent from the lower level are received at the upper level, where they are filtered by a robust data-driven DSE. The proposed hybrid framework does not depend on the centralized infrastructure of the control centers; thus it can be completely embedded into the wide-area measurement systems. This feature will ultimately facilitate the placement of hierarchical decentralized control schemes at the phasor data concentrator locations. Also, the network model is not necessary; thus, a topology processor is not required. Finally, there is no assumption on the dynamics of the electric loads. The proposed framework is tested on the 2,000-bus synthetic Texas system and shown to be capable of reconstructing the dynamic states of the generators with high accuracy, and of forecasting in the advent of missing data.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Classification of animal sounds in a hyperdiverse rainforest using convolutional neural networks with data augmentation

To protect tropical forest biodiversity, we need to be able to detect it reliably, cheaply, and at scale. Automated detection of sound producing animals from passively recorded soundscapes via machine-learning approaches is a promising technique towards this goal, but it is constrained by the necessity of large training data sets. Using soundscapes from a tropical forest in Borneo and a Convolutional Neural Network model (CNN), we investigate i) the minimum viable training data set size for accurate prediction of call types (‘sonotypes’), and ii) the extent to which data augmentation and transfer learning can overcome the issue of small and imbalanced training data sets. We found that even relatively high sample sizes (>80 per sonotype) lead to mediocre accuracy, which however improved significantly with data augmentation and transfer learning, including at extremely small sample sizes (3 per sonotype), regardless of taxonomic group or call characteristics. Neither transfer learning nor data augmentation alone achieved high accuracy. Our results suggest that transfer learning and data augmentation could make the use of CNNs to classify species’ vocalizations feasible even for small soundscape-based projects with many rare species. Retraining our open-source model requires only basic programming skills which makes it possible for individual conservation initiatives to match their local context, in order to enable more evidence-informed management of biodiversity.

54 ENVIRONMENTAL SCIENCES↗

Inference of cell type-specific gene regulatory networks on cell lineages from single cell omic datasets

Abstract Cell type-specific gene expression patterns are outputs of transcriptional gene regulatory networks (GRNs) that connect transcription factors and signaling proteins to target genes. Single-cell technologies such as single cell RNA-sequencing (scRNA-seq) and single cell Assay for Transposase-Accessible Chromatin using sequencing (scATAC-seq), can examine cell-type specific gene regulation at unprecedented detail. However, current approaches to infer cell type-specific GRNs are limited in their ability to integrate scRNA-seq and scATAC-seq measurements and to model network dynamics on a cell lineage. To address this challenge, we have developed single-cell Multi-Task Network Inference (scMTNI), a multi-task learning framework to infer the GRN for each cell type on a lineage from scRNA-seq and scATAC-seq data. Using simulated and real datasets, we show that scMTNI is a broadly applicable framework for linear and branching lineages that accurately infers GRN dynamics and identifies key regulators of fate transitions for diverse processes such as cellular reprogramming and differentiation.

59 BASIC BIOLOGICAL SCIENCES↗

Learning random networks for compression of still and moving images

Image compression for both still and moving images is an extremely important area of investigation, with numerous applications to videoconferencing, interactive education, home entertainment, and potential applications to earth observations, medical imaging, digital libraries, and many other areas. We describe work on a neural network methodology to compress/decompress still and moving images. We use the 'point-process' type neural network model which is closer to biophysical reality than standard models, and yet is mathematically much more tractable. We currently achieve compression ratios of the order of 120:1 for moving grey-level images, based on a combination of motion detection and compression. The observed signal-to-noise ratio varies from values above 25 to more than 35. The method is computationally fast so that compression and decompression can be carried out in real-time. It uses the adaptive capabilities of a set of neural networks so as to select varying compression ratios in real-time as a function of quality achieved. It also uses a motion detector which will avoid retransmitting portions of the image which have varied little from the previous frame. Further improvements can be achieved by using on-line learning during compression, and by appropriate compensation of nonlinearities in the compression/decompression scheme. We expect to go well beyond the 250:1 compression level for color images with good quality levels.

Gelenbe, Erol↗

Quantitative Analysis of the Drosophila Segmentation Regulatory Network Using Pattern Generating Potentials

Cis-regulatory modules that drive precise spatial-temporal patterns of gene expression are central to the process of metazoan development. We describe a new computational strategy to annotate genomic sequences based on their ‘‘pattern generating potential’’ and to produce quantitative descriptions of transcriptional regulatory networks at the level of individual protein-module interactions. We use this approach to convert the qualitative understanding of interactions that regulate Drosophila segmentation into a network model in which a confidence value is associated with each transcription factor-module interaction. Sequence information from multiple Drosophila species is integrated with transcription factor binding specificities to determine conserved binding site frequencies across the genome. These binding site profiles are combined with transcription factor expression information to create a model to predict module activity patterns. This model is used to scan genomic sequences for the potential to generate all or part of the expression pattern of a nearby gene, obtained from available gene expression databases. Interactions between individual transcription factors and modules are inferred by a statistical method to quantify a factor’s contribution to the module’s pattern generating potential. We use these pattern generating potentials to systematically describe the location and function of known and novel cis-regulatory modules in the segmentation network, identifying many examples of modules predicted to have overlapping expression activities. Surprisingly, conserved transcription factor binding site frequencies were as effective as experimental measurements of occupancy in predicting module expression patterns or factor-module interactions. Thus, unlike previous module prediction methods, this method predicts not only the location of modules but also their spatial activity pattern and the factors that directly determine this pattern. As databases of transcription factor specificities and in vivo gene expression patterns grow, analysis of pattern generating potentials provides a general method to decode transcriptional regulatory sequences and networks.

59 BASIC BIOLOGICAL SCIENCES↗

Genome-Based Metabolic Mapping and 13 C Flux Analysis Reveal Systematic Properties of an Oleaginous Microalga Chlorella protothecoides

We report that integrated and genome-based flux balance analysis, metabolomics, and 13 C-label profiling of phototrophic and heterotrophic metabolism in Chlorella protothecoides, an oleaginous green alga for biofuel. The green alga Chlorella protothecoides, capable of autotrophic and heterotrophic growth with rapid lipid synthesis, is a promising candidate for biofuel production. Based on the newly available genome knowledge of the alga, we reconstructed the compartmentalized metabolic network consisting of 272 metabolic reactions, 270 enzymes, and 461 encoding genes and simulated the growth in different cultivation conditions with flux balance analysis. Phenotype-phase plane analysis shows conditions achieving theoretical maximum of the biomass and corresponding fatty acid-producing rate for phototrophic cells (the ratio of photon uptake rate to CO 2 uptake rate equals 8.4) and heterotrophic ones (the glucose uptake rate to O 2 consumption rate reaches 2.4), respectively. Isotope-assisted liquid chromatography-mass spectrometry/mass spectrometry reveals higher metabolite concentrations in the glycolytic pathway and the tricarboxylic acid cycle in heterotrophic cells compared with autotrophic cells. We also observed enhanced levels of ATP, nicotinamide adenine dinucleotide (phosphate), reduced, acetyl-Coenzyme A, and malonyl-Coenzyme A in heterotrophic cells consistently, consistent with a strong activity of lipid synthesis. To profile the flux map in experimental conditions, we applied nonstationary 13 C metabolic flux analysis as a complementing strategy to flux balance analysis. We found that the result reveals negligible photorespiratory fluxes and a metabolically low active tricarboxylic acid cycle in phototrophic C. protothecoides. In comparison, high throughput of amphibolic reactions and the tricarboxylic acid cycle with no glyoxylate shunt activities were measured for heterotrophic cells. Lastly, taken together, the metabolic network modeling assisted by experimental metabolomics and 13 C labeling better our understanding on global metabolism of oleaginous alga, paving the way to the systematic engineering of the microalga for biofuel production.

09 BIOMASS FUELS↗

Do genome-scale models need exact solvers or clearer standards?

Constraint-based analysis of genome-scale models (GEMs) arose shortly after the first genome sequences became available. As numerous reviews of the field show, this approach and methodology has proven to be successful in studying a wide range of biological phenomena (McCloskey et al, 2013; Bordbar et al, 2014). However, efforts to expand the user base are impeded by hurdles in correctly formulating these problems to obtain numerical solutions. In particular, in a study entitled “An exact arithmetic toolbox for a consistent and reproducible structural analysis of metabolic network models” (Chindelevitch et al, 2014), the authors apply an exact solver to 88 genome-scale constraint-based models of metabolism. The authors claim that COBRA calculations (Orth et al, 2010) are inconsistent with their results and that many published and actively used (Lee et al, 2007; McCloskey et al, 2013) genome-scale models do support cellular growth in existing studies only because of numerical errors. They base these broad claims on two observations: (i) three reconstructions ( i AF1260, i IT341, and i NJ661) compute feasibly in COBRA, but are infeasible when exact numerical algorithms are used by their software (entitled MONGOOSE); (ii) linear programs generated by MONGOOSE for i IT341 were submitted to the NEOS Server (a Web site that runs linear programs through various solvers) and gave inconsistent results. They further claim that a large percentage of these COBRA models are actually unable to produce biomass flux. Here, we demonstrate that the claims made by Chindelevitch et al (2014) stem from an incorrect parsing of models from files rather than actual problems with numerical error or COBRA computations.

59 BASIC BIOLOGICAL SCIENCES↗