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 19 records

Orthogonal patterns in binary neural networks

A binary neural network that stores only mutually orthogonal patterns is shown to converge, when probed by any pattern, to a pattern in the memory space, i.e., the space spanned by the stored patterns. The latter are shown to be the only members of the memory space under a certain coding condition, which allows maximum storage of M=(2N) sup 0.5 patterns, where N is the number of neurons. The stored patterns are shown to have basins of attraction of radius N/(2M), within which errors are corrected with probability 1 in a single update cycle. When the probe falls outside these regions, the error correction capability can still be increased to 1 by repeatedly running the network with the same probe.

Baram, Yoram↗

Hyperparameter Optimization in Binary Communication Networks for Neuromorphic Deployment

Training neural networks for neuromorphic deployment is non-trivial. There have been a variety of approaches proposed to adapt back-propagation or back-propagation-like algorithms appropriate for training. Considering that these networks often have very different performance characteristics than traditional neural networks, it is often unclear how to set either the network topology or the hyperparameters to achieve optimal performance. In this work, we introduce a Bayesian approach for optimizing the hyperparameters of an algorithm for training binary communication networks that can be deployed to neuromorphic hardware. We show that by optimizing the hyperparameters on this algorithm for each dataset, we can achieve improvements in accuracy over the previous state-of-the-art for this algorithm on each dataset (by up to 15 percent). This jump in performance continues to emphasize the potential when converting traditional neural networks to binary communication applicable to neuromorphic hardware.

Parsa, Maryam↗

GAAF: Searching Activation Functions for Binary Neural Networks Through Genetic Algorithm

Binary neural networks (BNNs) show promising utilization in cost and power-restricted domains such as edge devices and mobile systems. This is due to its significantly less computation and storage demand, but at the cost of degraded performance. To close the accuracy gap, in this paper we propose to add a complementary activation function (AF) ahead of the sign based binarization, and rely on the genetic algorithm (GA) to automatically search for the ideal AFs. These AFs can help extract extra information from the input data in the forward pass, while allowing improved gradient approximation in the backward pass. Fifteen novel AFs are identified through our GA-based search, while most of them show improved performance (up to 2.54% on ImageNet) when testing on different datasets and network models. Interestingly, periodic functions are identified as a key component for most of the discovered AFs, which rarely exist in human designed AFs. Our method offers a novel approach for designing general and application-specific BNN architecture.

59 BASIC BIOLOGICAL SCIENCES↗

Orthogonal Patterns In A Binary Neural Network

Report presents some recent developments in theory of binary neural networks. Subject matter relevant to associate (content-addressable) memories and to recognition of patterns - both of considerable importance in advancement of robotics and artificial intelligence. When probed by any pattern, network converges to one of stored patterns.

Baram, Yoram↗

Classification of multispectral image data by the Binary Diamond neural network and by nonparametric, pixel-by-pixel methods

The classification of multispectral image data obtained from satellites has become an important tool for generating ground cover maps. This study deals with the application of nonparametric pixel-by-pixel classification methods in the classification of pixels, based on their multispectral data. A new neural network, the Binary Diamond, is introduced, and its performance is compared with a nearest neighbor algorithm and a back-propagation network. The Binary Diamond is a multilayer, feed-forward neural network, which learns from examples in unsupervised, 'one-shot' mode. It recruits its neurons according to the actual training set, as it learns. The comparisons of the algorithms were done by using a realistic data base, consisting of approximately 90,000 Landsat 4 Thematic Mapper pixels. The Binary Diamond and the nearest neighbor performances were close, with some advantages to the Binary Diamond. The performance of the back-propagation network lagged behind. An efficient nearest neighbor algorithm, the binned nearest neighbor, is described. Ways for improving the performances, such as merging categories, and analyzing nonboundary pixels, are addressed and evaluated.

Salu, Yehuda↗

BitGNN: Unlocking the Performance Potential of Binary Graph Neural Networks on GPUs

Graph Neural Networks (GNNs) have shown compelling results in many graph-based learning tasks. They are, however, time-consuming. Recent work has shown a promising direction in improving GNN speed and shrinking the size — network binarization, which binarizes network values and operations. Prior work, however, mainly focused on algorithm designs, leaving it open on how to fully materialize the performance potential. This work fills the gap by proposing techniques to best map binary GNNs and their computations to fit the nature of bit manipulations, optimizations and algorithms to maximize BSpMM kernel efficiency, and solutions to other factors influencing the end-to-end time on GPUs. Results on real-world graphs show that the proposed techniques outperform state of-the-art binary GNN implementations by 21-67× with little accuracy loss.

Chen, Jou-An↗

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↗

Ground states of partially connected binary neural networks

Neural networks defined by outer products of vectors over (-1, 0, 1) are considered. Patterns over (-1, 0, 1) define by their outer products partially connected neural networks consisting of internally strongly connected, externally weakly connected subnetworks. Subpatterns over (-1, 1) define subnetworks, and their combinations that agree in the common bits define permissible words. It is shown that the permissible words are locally stable states of the network, provided that each of the subnetworks stores mutually orthogonal subwords, or, at most, two subwords. It is also shown that when each of the subnetworks stores two mutually orthogonal binary subwords at most, the permissible words, defined as the combinations of the subwords (one corresponding to each subnetwork), that agree in their common bits are the unique ground states of the associated energy function.

Baram, Yoram↗

On the stability, storage capacity, and design of nonlinear continuous neural networks

The stability, capacity, and design of a nonlinear continuous neural network are analyzed. Sufficient conditions for existence and asymptotic stability of the network's equilibria are reduced to a set of piecewise-linear inequality relations that can be solved by a feedforward binary network, or by methods such as Fourier elimination. The stability and capacity of the network is characterized by the post synaptic firing rate function. An N-neuron network with sigmoidal firing function is shown to have up to 3N equilibrium points. This offers a higher capacity than the (0.1-0.2)N obtained in the binary Hopfield network. Moreover, it is shown that by a proper selection of the postsynaptic firing rate function, one can significantly extend the capacity storage of the network.

Guez, Allon↗

The Interplay of Binary and Quantitative Structure on the Stability of Mutualistic Networks

Synopsis Understanding how the structure of biological systems impacts their resilience (broadly defined) is a recurring question across multiple levels of biological organization. In ecology, considerable effort has been devoted to understanding how the structure of interactions between species in ecological networks is linked to different broad resilience outcomes, especially local stability. Still, nearly all of that work has focused on interaction structure in presence-absence terms and has not investigated quantitative structure, i.e., the arrangement of interaction strengths in ecological networks. We investigated how the interplay between binary and quantitative structure impacts stability in mutualistic interaction networks (those in which species interactions are mutually beneficial), using community matrix approaches. We additionally examined the effects of network complexity and within-guild competition for context. In terms of structure, we focused on understanding the stability impacts of nestedness, a structure in which more-specialized species interact with smaller subsets of the same species that more-generalized species interact with. Most mutualistic networks in nature display binary nestedness, which is puzzling because both binary and quantitative nestedness are known to be destabilizing on their own. We found that quantitative network structure has important consequences for local stability. In more-complex networks, binary-nested structures were the most stable configurations, depending on the quantitative structures, but which quantitative structure was stabilizing depended on network complexity and competitive context. As complexity increases and in the absence of within-guild competition, the most stable configurations have a nested binary structure with a complementary (i.e., anti-nested) quantitative structure. In the presence of within-guild competition, however, the most stable networks are those with a nested binary structure and a nested quantitative structure. In other words, the impact of interaction overlap on community persistence is dependent on the competitive context. These results help to explain the prevalence of binary-nested structures in nature and underscore the need for future empirical work on quantitative structure.

Zoology↗

Early Information Parameter-Set Analysis for Satellite Close Approaches using Machine Learning

Understanding orbital mechanics is essential in space flight and navigation applications, and leveraging modern force models for flight path projection remains an important aspect in space mission design and operation. However, force models do not capture all the dynamics or perturbations in the space environment and thus are subject to errors in predicting the state vectors. The further out the predicted miss distance between spacecraft is from the time of closest approach (TCA), the larger the propagated errors in the predicted miss distance at TCA is. The dependency on these force models for spacecraft flight state prediction calls for a more reliable method that can quantify, or even reduce, these propagated errors. With recent advances in the field artificial intelligence, specifically in machine and deep learning algorithms, a model that implements these approaches can improve on the modern force model approach. The goal for this work is to provide an early-information decision-making threshold, in order to prioritize risk assessment implementation, given the ongoing increase of space objects. In analyzing the relationship of several parameters from conjunction data messages(CDMs) and solar information, early information becomes viable in miss distance prediction with unsupervised learning techniques, which learn the parameters that are linked together with miss distance and probability of collision (Pc) variables. Another approach implemented for identifying relationships within CDMs is supervised learning, in which a shallow neural network binary classifier learns to distinguish events with Pc values¡108. These parameters detected in the unsupervised process are then applied to a regression neural network, which predicts the miss distance at TCA for a specific event within a given uncertainty bound. For the regression neural network, a Long Short Term Memory (LSTM) neural network is implemented, which yields memory about each time step in an event. Using an LSTM network, the model learns to predict miss distance within 0.2km of the value measured at TCA. Although there is a limited amount of "close miss" data to train a network, the network learns to associate parameters, like large energy dissipation rates with the secondary object, with an elevated Pc

Brianna I. Robertson↗

Designing Glass and Crystalline Phases of Metal–Bis(acetamide) Networks to Promote High Optical Contrast

Owing to their high tunability and predictable structures, metal–organic materials offer a powerful platform to study glass formation and crystallization processes and to design glasses with unique properties. In this work, we report a novel series of glass-forming metal–ethylenebis(acetamide) networks that undergo reversible glass and crystallization transitions below 200 °C. The glass-transition temperatures, crystallization kinetics, and glass stability of these materials are readily tunable, either by synthetic modification or by liquid-phase blending, to form binary glasses. Pair distribution function (PDF) analysis reveals extended structural correlations in both single and binary metal–bis(acetamide) glasses and highlights the important role of metal–metal correlations during structural evolution across glass–crystal transitions. Notably, the glass and crystalline phases of a Co–ethylenebis(acetamide) binary network feature a large reflectivity contrast ratio of 4.8 that results from changes in the local coordination environment around Co centers. These results provide new insights into glass–crystal transitions in metal–organic materials and have exciting implications for optical switching, rewritable data storage, and functional glass ceramics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

On the capacity of ternary Hebbian networks

Networks of ternary neurons storing random vectors over the set -1,0,1 by the so-called Hebbian rule are considered. It is shown that the maximal number of stored patterns that are equilibrium states of the network with probability tending to one as N tends to infinity is at least on the order of (N exp 2-1/alpha)/K, where N is the number of neurons, K is the number of nonzero elements in a pattern, and t = alpha x K, alpha between 1/2 and 1, is the threshold in the neuron function. While, for small K, this bound is similar to that obtained for fully connected binary networks, the number of interneural connections required in the ternary case is considerably smaller. Similar bounds, incorporating error probabilities, are shown to guarantee, in the same probabilistic sense, the correction of errors in the nonzero elements and in the location of these elements.

Baram, Yoram↗