Search NASA⌕ Search

SEARCH · Search NASA

Results for “binary network”

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

Efficient human activity recognition with spatio-temporal spiking neural networks

In this study, we explore Human Activity Recognition (HAR), a task that aims to predict individuals' daily activities utilizing time series data obtained from wearable sensors for health-related applications. Although recent research has predominantly employed end-to-end Artificial Neural Networks (ANNs) for feature extraction and classification in HAR, these approaches impose a substantial computational load on wearable devices and exhibit limitations in temporal feature extraction due to their activation functions. To address these challenges, we propose the application of Spiking Neural Networks (SNNs), an architecture inspired by the characteristics of biological neurons, to HAR tasks. SNNs accumulate input activation as presynaptic potential charges and generate a binary spike upon surpassing a predetermined threshold. This unique property facilitates spatio-temporal feature extraction and confers the advantage of low-power computation attributable to binary spikes. We conduct rigorous experiments on three distinct HAR datasets using SNNs, demonstrating that our approach attains competitive or superior performance relative to ANNs, while concurrently reducing energy consumption by up to 94%.

60 APPLIED LIFE SCIENCES↗

Identifying the nature of the QCD transition in heavy-ion collisions with deep learning

In this proceeding, we review our recent work using deep convolutional neural network (CNN) to identify the nature of the QCD transition in a hybrid modeling of heavy-ion collisions. Within this hybrid model, a viscous hydrodynamic model is coupled with a hadronic cascade “after-burner”. As a binary classification setup, we employ two different types of equations of state (EoS) of the hot medium in the hydrodynamic evolution. The resulting final-state pion spectra in the transverse momentum and azimuthal angle plane are fed to the neural network as the input data in order to distinguish different EoS. To probe the effects of the fluctuations in the event-by-event spectra, we explore different scenarios for the input data and make a comparison in a systematic way. We observe a clear hierarchy in the predictive power when the network is fed with the event-by-event, cascade-coarse-grained and event-fine-averaged spectra. The carefully-trained neural network can extract high-level features from pion spectra to identify the nature of the QCD transition in a realistic simulation scenario.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Impact of O4 future detections on the determination of the dense matter equations of state

Here, in view of the next LIGO-Virgo-KAGRA Observing period O4 (to start in Spring 2023), we address the question of the ability of the interferometers network to discriminate among different neutron stars equation of states better than what was possible with the observation of the binary neutron stars merger GW170817. We show that the observation of an event similar to GW170817 during O4 would allow us to resolve the dimensionless effective tidal deformability $\tilde{Λ}$ within an uncertainty 7 times better than the one obtained in O2. Thanks to the expected increase in sensitivities, we show that any GW170817-like single-event within a distance of 100 Mpc would imply significantly improved constraints of the neutron stars equations of state. We also illustrate the important impact of the noise in the analysis of the signal, showing how it can impact the effective tidal deformability probability density function for large signal-to-noise ratio.

79 ASTRONOMY AND ASTROPHYSICS↗

A Bright Short Period M-M Eclipsing Binary from the KELT Survey: Magnetic Activity and the Mass–Radius Relationship for M Dwarfs

We report the discovery of KELT J041621-620046, a moderately bright (J ∼ 10.2) M-dwarf eclipsing binary system at a distance of 39 ± 3 pc. KELT J041621-620046 was first identified as an eclipsing binary using observations from the Kilodegree Extremely Little Telescope (KELT) survey. The system has a short orbital period of ∼1.11 days and consists of components with M{sub 1}=0.447{sub +0.052}{sup −0.047} M{sub ⊙} and M{sub 2}=0.399{sub +0.046}{sup −0.042} M{sub ⊙} in nearly circular orbits. The radii of the two stars are R{sub 1}=0.540{sub +0.034}{sup −0.032} R{sub ⊙} and R{sub 2}=0.453±0.017 R{sub ⊙}. Full system and orbital properties were determined (to ∼10% error) by conducting an EBOP (Eclipsing Binary Orbit Program) global modeling of the high precision photometric and spectroscopic observations obtained by the KELT Follow-up Network. Each star is larger by 17%–28% and cooler by 4%–10% than predicted by standard (non-magnetic) stellar models. Strong Hα emission indicates chromospheric activity in both stars. The observed radii and temperature discrepancies for both components are more consistent with those predicted by empirical relations that account for convective suppression due to magnetic activity.

79 ASTRONOMY AND ASTROPHYSICS↗

Quantum cluster equilibrium model of N-methylformamide–water binary mixtures

The established quantum cluster equilibrium (QCE) approach is refined and applied to N-methylformamide (NMF) and its aqueous solution. The QCE method is split into two iterative cycles: one which converges to the liquid phase solution of the QCE equations and another which yields the gas phase. By comparing Gibbs energies, the thermodynamically stable phase at a given temperature and pressure is then chosen. The new methodology avoids metastable solutions and allows a different treatment of the mean-field interactions within the gas and liquid phases. These changes are of crucial importance for the treatment of binary mixtures. For the first time in a QCE study, the cis-trans-isomerism of a species (NMF) is explicitly considered. Cluster geometries and frequencies are calculated using density functional theory (DFT) and complementary coupled cluster single point energies are used to benchmark the DFT results. Independent of the selected quantum-chemical method, a large set of clusters is required for an accurate thermodynamic description of the binary mixture. The liquid phase of neat NMF is found to be dominated by the cyclic trans-NMF pentamer, which can be interpreted as a linear trimer that is stabilized by explicit solvation of two further NMF molecules. This cluster reflects the known hydrogen bond network preferences of neat NMF.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Experimental and Theoretical Insights into the Structure of Tellurium Chloride Glasses

Here, the structure of the binary chalcohalide glasses Te 1-x CL x (0.35 <= x <= 0.65) is considered by combining experimental and theoretical results. The structural network properties are influenced by a competition between ionic and covalent bonding in such glasses. At first, a focus is placed on the detailed information available by using the complementary high-energy X-ray and the neutron diffractions in both the reciprocal and real spaces. The main characteristic suggested by the structure factors S(Q) concerns the presence of three length scales in the intermediate range order. The total correlation function T(r) lets us also suppose that the structure of these glasses is more complicated than Te-chain fragments with terminal Cl as demonstrated in crystalline Te 3 Cl 2 . Molecular dynamics simulations were subsequently performed on Te 3 Cl 2 and Te 2 Cl 3 , and coupled with the experimental data, a highly reticulated network of chalcogen atoms, with a fair amount of chlorine atoms bonded in a bridging mode, is proposed. The simulations clearly lead to a glass description that differs markedly from the simple structural model based on only Te atom chains and terminal Cl atoms. Solidstate NMR experiments and NMR parameters calculations allowed validation of the presence of Te highly coordinated with chlorine in these glasses.

36 MATERIALS SCIENCE↗

Digital carrier demodulation for the DSN Advanced Receiver

The digital in-phase and quadrature (I and Q) carrier demodulation for the Deep Space Network's (DSN) Advanced Receiver is described and analyzed. The system provides coherent demodulation for a variety of modulation formats including Binary Phase Shift Keying (BPSK), BPSK with a carrier residual, Quadrature Phase Shift Keying (QPSK), Offset-QPSK (OQPSK), and Minimum Shift Keying (MSK). The focus is on the theory and hardware design of the halfband filters which are the integral part of the demodulator. The underlying theory of the filters is summarized, a breadboard hardware design is described, and a VLSI implementation is proposed which significantly decreases the hardware. A second important problem analyzed is DC-offset in the demodulator. This is a serious problem which translates into bias error in the residual carrier phase detector. The dynamic range of the complex mixer is analyzed using a probabilistic approach. It is deduced that the resulting static phase error is less than 0.2 deg when the ratio of carrier power to noise power in the demodulator input bandwidth is -72 dB or higher. Thus, the static phase error is negligible at a carrier power to noise spectral density of 0 dB-Hz for a 15 MHz bandwidth demodulator.

Sadr, R.↗

Resistive synaptic interconnects for electronic neural networks

The use of the alpha-Ge(1-x):Al(x) and alpha-Ge(1-x):Cu(x) alloys and Pt/Al2O3 cermet thin films as resistive interconnects for binary synaptic memory arrays is evaluated. The fabrication of the 10-20 microns long, 10 microns wide, and 0.1 micron thick interconnects from the alloys and cermet is described. The current-voltage and switching characteristics of the as-deposited films and the patterned test structure are studied. The resistivity, uniformity, stability, and compatibility of the interconnects are examined. It is observed that alpha-Ge(1-x):Cu(x) alloys have a wide resistivity range and low temperature coefficients of resistance; however, their long-term stability is limited due to their low crystallization temperature. It is detected that the alpha-Ge(1-x):Al(x) alloys have higher crystallization temperatures and their resistivity is not greatly affected by large changes in metal content. The Pt/Al2O3 samples display excellent stability, easy fabrication, and control of resistivity with metal content.

Lamb, J. L.↗

Neural network based feed-forward high density associative memory

A novel thin film approach to neural-network-based high-density associative memory is described. The information is stored locally in a memory matrix of passive, nonvolatile, binary connection elements with a potential to achieve a storage density of 10 to the 9th bits/sq cm. Microswitches based on memory switching in thin film hydrogenated amorphous silicon, and alternatively in manganese oxide, have been used as programmable read-only memory elements. Low-energy switching has been ascertained in both these materials. Fabrication and testing of memory matrix is described. High-speed associative recall approaching 10 to the 7th bits/sec and high storage capacity in such a connection matrix memory system is also described.

Daud, T.↗

Extending Power of Nature from Binary Problems to Real-Valued Graph Learning in Real World

Nature performs complex computations constantly at clearly lower cost and higher performance than digital computers. It is crucial to understand how to harness the unique computational power of nature in Machine Learning (ML). In the past decade, besides the development of Neural Networks (NNs), the community has also relentlessly explored nature-powered ML paradigms. Although most of them are still predominantly theoretical, a new practical paradigm enabled by the recent advent of CMOS-compatible room-temperature nature-based computers has emerged. By harnessing the nature's power of entropy increase, this paradigm can solve binary learning problems delivering immense speedup and energy savings compared with NNs, while maintaining comparable accuracy. Regrettably, its values to the real world are highly constrained by its binary nature. A clear pathway to its extension to real-valued problems remains elusive. This paper aims to unleash this pathway by proposing a novel end-to-end Nature-Powered Graph Learning (NP-GL) framework. Specifically, through a three-dimensional co-design, NP-GL can leverage the nature's power of entropy increase to efficiently solve real-valued graph learning problems. Experimental results across 4 real-world applications with 6 datasets demonstrate that NP-GL delivers, on average, 6970X speedup and 10^5x energy consumption reduction with comparable or even higher accuracy than Graph Neural Networks (GNNs).

artificial intelligence↗

O3BNN: An Out-Of-Order Architecture for High-Performance Binarized Neural Network Inference with Fine-Grained Pruning

In this work, we demonstrate that the highly-condensed BNN model can be shrunk significantly further by dynamically pruning irregular redundant edges. Based on two new observations on BNN-specific properties, our out-of-order (OoO) architecture – O3BNN, can curtail remaining edge evaluation in cases where the binary output of a neuron can be determined early. Similar to Instruction-Level-Parallelism (ILP), these fine-grained, irregular, runtime pruning opportunities are traditionally presumed to be di cult to exploit. We evaluate our design on an FPGA platform using three well-known networks, including VggNet-16, AlexNet for ImageNet, and a VGG-like network for Cifar-10. Results show that our out-of-order approach can prune 27%, 16%, and 42% of the operations for the three networks respectively, without any accuracy loss, leading to at least 1.7×, 1.5×, and 2.1× speedups over state-of-the-art implementations on FPGA/GPU/CPU BNN implementations. Our approach is inference runtime pruning, so no retrain or fine-tuning is needed. We demonstrate our design on an FPGA platform. However, this is only for showcasing the method; the approach does not rely on any FPGA-specific features and can thus be adopted by other devices as well.

Geng, Tong↗

Grain boundary slip transfer classification and metric selection with artificial neural networks

An artificial neural network is used to evaluate the effectiveness of six metrics and their combinations to assess whether slip transfers across grain boundaries in coarse-grained oligocrystalline Al foils. This approach extends the one- or two-dimensional projections formerly applied to analyze slip transfer. The accuracy of this binary classification reaches around 87% for the best single metric and around 90% when considering two or more metrics simultaneously. Here, the results suggest slip transfer mostly depends on the geometric relationship between grains. Training a double-layer network having 10 nodes per hidden layer with 40 measurements is sufficient to render the maximum accuracy.

36 MATERIALS SCIENCE↗

Tension-free Dirac strings and steered magnetic charges in 3D artificial spin ice

3D nano-architectures presents a new paradigm in modern condensed matter physics with numerous applications in photonics, biomedicine, and spintronics. They are promising for the realization of 3D magnetic nano-networks for ultra-fast and low-energy data storage. Frustration in these systems can lead to magnetic charges or magnetic monopoles, which can function as mobile, binary information carriers. However, Dirac strings in 2D artificial spin ices bind magnetic charges, while 3D dipolar counterparts require cryogenic temperatures for their stability. Here, we present a micromagnetic study of a highly frustrated 3D artificial spin ice harboring tension-free Dirac strings with unbound magnetic charges at room temperature. We use micromagnetic simulations to demonstrate that the mobility threshold for magnetic charges is by 2 eV lower than their unbinding energy. By applying global magnetic fields, we steer magnetic charges in a given direction omitting unintended switchings. The introduced system paves the way toward 3D magnetic networks for data transport and storage.

36 MATERIALS SCIENCE↗

BINARY COMPACT OBJECT COALESCENCE RATES: THE ROLE OF ELLIPTICAL GALAXIES

In this paper, we estimate binary compact object merger detection rates for LIGO, including the potentially significant contribution from binaries that are produced in elliptical galaxies near the epoch of peak star formation. Specifically, we convolve hundreds of model realizations of elliptical- and spiral-galaxy population syntheses with a model for elliptical- and spiral-galaxy star formation history as a function of redshift. Our results favor local merger rate densities of 4 x 10{sup -3} Mpc{sup -3} Myr{sup -1} for binary black holes (BHs), 3 x 10{sup -2} Mpc{sup -3} Myr{sup -1} for binary neutron stars (NSs), and 10{sup -2} Mpc{sup -3} Myr{sup -1} for BH-NS binaries. We find that mergers in elliptical galaxies are a significant fraction of our total estimate for BH-BH and BH-NS detection rates; NS-NS detection rates are likely dominated by the contribution from spiral galaxies. Limiting attention to elliptical-galaxy plus only those spiral-galaxy models that reproduce current observations of Galactic NS-NS, we find slightly higher rates for NS-NS and largely similar ranges for BH-NS and BH-BH binaries. Assuming a detection signal-to-noise ratio threshold of 8 for a single detector (in practice as part of a network, to reduce its noise), corresponding to radii D {sub bns} of the effective volume inside of which a single LIGO detector could observe the inspiral of two 1.4 M {sub sun} NSs of 14 Mpc and 197 Mpc, for initial and advanced LIGO, we find event rates of any merger type of 2.9 x 10{sup -2}-0.46 and 25-400 yr{sup -1} (at 90% confidence level), respectively. We also find that the probability P {sub detect} of detecting one or more mergers with this single detector can be approximated by (1) P {sub detect} {approx_equal} 0.4 + 0.5 log(T/0.01 yr), assuming D {sub bns} = 197 Mpc and it operates for T yr, for T between 2 days and 0.1 yr, or by (2) P {sub detect} {approx_equal} 0.5 + 1.5 log(D {sub bns}/32 Mpc), for 1 yr of operation and for D {sub bns} between 20 and 70 Mpc.

79 ASTRONOMY AND ASTROPHYSICS↗

Predictive Modeling for Differential Diagnosis and Mortality Risk Assessment

The prevalence of electronic health record (EHR) systems has brought prodigious biomedical informatics opportunity. Automated machine learning methods can effectively utilize such data and have become common tools for healthcare predictive modeling. Researches in medical informatics have explored the potential of deep learning and classical models in emergent care scenarios. In particular, predicting differential diagnoses for admissions have proven useful in decreasing unnecessary lab tests and improving inpatient triage decision-making. Moreover, identification of high-risk patients for in-hospital mortality is vitally important to maximize allocation of medical resources.The Medical Information Mart for Intensive Care (MIMIC-III) database, containing de-identified critical care inpatient was used in our study. This data set captures hospital patient laboratory measurements, pharmacologic prescriptions, diagnostic data and procedure event recordings. When considering adult patients and discounting admissions with ICU length of stay less than 24 hours, there were 37,787 unique admissions and 30,414 total patients. We examined the top 25 most prevalent ICD-9 group-level disease specificities in MIMIC-III using a multi-label classification model. In-hospital mortality was modeled as binary classification with 4,155 (13%) adult patients that expired, of which 3,138 (75.5%) were in the ICU setting. The metrics AUC, F1 score, sensitivity and specificity values calculated for each disease label measured prediction performance.The usage of ICD-9 group codes reduced feature dimension from 14,567 to 942 and greatly improved distribution of patient diagnostic categories. Disease temporal patterns were captured by considering the most frequently sampled 6 vital signs and 13 laboratory values. Missing data were imputed at each time-stamp. Time-series raw hourly average values were converted into 5 summary features (mean, standard deviation, number of observations, min & max values). Patient demographic variables such as age, gender, marital status and ethnicity were also factored into the modeling. Choi et al showed that contextual embedding of medical data, diagnostic and procedural codes alone can predict future diagnoses with sensitivity as high as 0.79. We utilized an embedding technique called word2vec which allowed sparse representations of medical history to be transformed into dense word vectors. The mappings captured contextual information by treating each admission as a sentence and learning the most likely neighboring words in a sliding window fashion. Binary and multi-label classification was achieved via collapse models, which do not consider temporal information, as well as recurrent neural networks with regularization, Softmax output layer activation together with categorical cross-entropy as the loss function.

US Army collaboration↗

Sandbox for Mac Malware v 1.0

This software is an analyzer for automated sandbox analysis of malware on the OS X operating system. It runs inside an OS X virtual machine to collect data about what happens when a given file is opened or run. As of August 2014, there was no sandbox software for Mac OS X malware, as it requires different methods from those used on the Windows OS (which most sandboxes are written for). This software adds OS X analysis capabilities to an existing open-source sandbox, Cuckoo Sandbox (http://cuckoosandbox.org/), which previously only worked for Windows. The analyzer itself can take many different types of files as input: the traditional Mach-O and FAT executables, .app files, zip files, Python scripts, Java archives, and web pages, as well as PDFs and other documents. While the file is running, the analyzer also simulates rudimentary human interaction with clicks and mouse movements in order to bypass the tests some malware use to see if they are being analyzed. The analyzer outputs several different kinds of data: function call traces, network captures, screenshots, and all created and modified files. This work also includes a static analysis Cuckoo module for Mach-O binary files. It extracts file structures, code library imports and exports, and signatures. This data can be used along with the analyzer results to create signatures for malware.

Walkup, Elizabeth↗

U.S. stock market interaction network as learned by the Boltzmann machine

Here, we study historical dynamics of joint equilibrium distribution of stock returns in the U.S. stock market using the Boltzmann distribution model being parametrized by external fields and pairwise couplings. Within Boltzmann learning framework for statistical inference, we analyze historical behavior of the parameters inferred using exact and approximate learning algorithms. Since the model and inference methods require use of binary variables, effect of this mapping of continuous returns to the discrete domain is studied. The presented results show that binarization preserves the correlation structure of the market. Properties of distributions of external fields and couplings as well as the market interaction network and industry sector clustering structure are studied for different historical dates and moving window sizes. We demonstrate that the observed positive heavy tail in distribution of couplings is related to the sparse clustering structure of the market. We also show that discrepancies between the model’s parameters might be used as a precursor of financial instabilities.

97 MATHEMATICS AND COMPUTING↗