An explicit form of the Mie phase matrix for multiple scattering calculations in the I, Q, U and V representation.
Explore the source record for details and available documents.
Kattawar, G. W.↗
SEARCH · Engineering Papers
Results for “matrix multiplication”
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.
Communication Lower Bounds and Optimal Algorithms for Multiple Tensor-Times-Matrix Computation
Multiple tensor-times-matrix (Multi-TTM) is a key computation in algorithms for computing and operating with the Tucker tensor decomposition, which is frequently used in multidimensional data analysis. Here, we establish communication lower bounds that determine how much data movement is required (under mild conditions) to perform the Multi-TTM computation in parallel. The crux of the proof relies on analytically solving a constrained, nonlinear optimization problem. We also present a parallel algorithm to perform this computation that organizes the processors into a logical grid with twice as many modes as the input tensor. We show that, with correct choices of grid dimensions, the communication cost of the algorithm attains the lower bounds and is therefore communication optimal. Finally, we show that our algorithm can significantly reduce communication compared to the straightforward approach of expressing the computation as a sequence of tensor-times-matrix operations when the input and output tensors vary greatly in size.
HBL-inequalities↗
Implementation of a Parameterized Interacting Multiple Model Filter on an FPGA for Satellite Communications
In a communications channel, the space environment between a spacecraft and an Earth ground station can potentially cause the loss of a data link or at least degrade its performance due to atmospheric effects, shadowing, multipath, or other impairments. In adaptive and coded modulation, the signal power level at the receiver can be used in order to choose a modulation-coding technique that maximizes throughput while meeting bit error rate (BER) and other performance requirements. It is the goal of this research to implement a generalized interacting multiple model (IMM) filter based on Kalman filters for improved received power estimation on software-dened radio (SDR) technology for satellite communications applications. The IMM filter has been implemented in Verilog consisting of a customizable bank of Kalman filters for choosing between performance and resource utilization. Each Kalman filter can be implemented using either solely a Schur complement module (for high area efficiency) or with Schur complement, matrix multiplication, and matrix addition modules (for high performance). These modules were simulated and synthesized for the Virtex II platform on the JPL Radio Experimenter Development System (EDS) at NASA Glenn Research Center. The results for simulation, synthesis, and hardware testing are presented.
cognitive radio↗
Binary operations on neuromorphic hardware with application to linear algebraic operations and stochastic equations
Abstract Non-von Neumann computational hardware, based on neuron-inspired, non-linear elements connected via linear, weighted synapses—so-called neuromorphic systems—is a viable computational substrate. Since neuromorphic systems have been shown to use less power than CPUs for many applications, they are of potential use in autonomous systems such as robots, drones, and satellites, for which power resources are at a premium. The power used by neuromorphic systems is approximately proportional to the number of spiking events produced by neurons on-chip. However, typical information encoding on these chips is in the form of firing rates that unarily encode information. That is, the number of spikes generated by a neuron is meant to be proportional to an encoded value used in a computation or algorithm. Unary encoding is less efficient (produces more spikes) than binary encoding. For this reason, here we present neuromorphic computational mechanisms for implementing binary two’s complement operations. We use the mechanisms to construct a neuromorphic, binary matrix multiplication algorithm that may be used as a primitive for linear differential equation integration, deep networks, and other standard calculations. We also construct a random walk circuit and apply it in Brownian motion simulations. We study how both algorithms scale in circuit size and iteration time.
97 MATHEMATICS AND COMPUTING↗
CCD data processor for maximum likelihood feature classification
The paper describes an advanced technology development which utilizes a high speed analog/binary CCD correlator to perform the matrix multiplications necessary to implement onboard feature classification. The matrix manipulation module uses the maximum likelihood classification algorithm assuming a Gaussian probability density function. The module will process 16 element multispectral vectors at rates in excess of 500 thousand multispectral vector elements per second. System design considerations for the optimum use of this module are discussed, test results from initial device fabrication runs are presented, and the performance in typical processing applications is described
Benz, H. F.↗
Analog VLSI neural network integrated circuits
Two analog very large scale integration (VLSI) vector matrix multiplier integrated circuit chips were designed, fabricated, and partially tested. They can perform both vector-matrix and matrix-matrix multiplication operations at high speeds. The 32 by 32 vector-matrix multiplier chip and the 128 by 64 vector-matrix multiplier chip were designed to perform 300 million and 3 billion multiplications per second, respectively. An additional circuit that has been developed is a continuous-time adaptive learning circuit. The performance achieved thus far for this circuit is an adaptivity of 28 dB at 300 KHz and 11 dB at 15 MHz. This circuit has demonstrated greater than two orders of magnitude higher frequency of operation than any previous adaptive learning circuit.
Kub, F. J.↗
Fatigue-life behavior and matrix fatigue crack spacing in unnotched SCS-6/Timetal 21S metal matrix composites
Fatigue tests of the SCS-6/Timetal 21S composite system were performed to characterize the fatigue behavior for unnotched conditions. The stress-life behavior of the unnotched (9/90)2s laminates was investigated for stress ratios of R = 0.1 and R = 0.3. The occurrence of matrix cracking was also examined in these specimens. This revealed multiple matrix crack initiation sites throughout the composite, as well as evenly spaced surface cracks along the length of the specimens. No difference in fatigue lives were observed for stress ratios of R = 0.1 and R = 0.3 when compared on a stress range basis. The unnotched SCS-6/Timetal 21S composites had shorter fatigue lives than the SCS-6/Ti-15-3 composites, however the neat Timetal 21S matrix material had a longer fatigue life than the neat Ti-15-3.
Ward, G. T.↗
Matrix differentiation formulas
A compact differentiation technique (without using indexes) is developed for scalar functions that depend on complex matrix arguments which are combined by operations of complex conjugation, transposition, addition, multiplication, matrix inversion and taking the direct product. The differentiation apparatus is developed in order to simplify the solution of extremum problems of scalar functions of matrix arguments.
Usikov, D. A.↗
Fault-tolerant high-performance multiplication: theory and practice
In this paper, we extend the theory and practice regarding algorithmic fault-tolerant matrix-matrix multiplication, C = AB, in a number of ways.
fault-tolerance BLAS fault-detection fault-correct↗
Use of Design of Experiments and Rule-Based Inference in Determining Neural Network Architectures for Loss of Control Detection
In this work, we describe methods for selecting the neural network architectures and input spaces to implement belief state inference on generic commercial transport aircraft. First, we highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). We present a structured method that can be used in a number of aerospace applications, to optimize the structure and training parameters of the CVAE for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from a NASA T-2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications. While theoretically, we could create a single CVAE with an input space the size of all measurable flight variables and environmental dynamics, it becomes intractable to use such a neural network in an in-situ intelligent multi-agent system. Using the recommendations from our case study, we introduce a technical approach for feasibly describing the belief space by (1) identifying significant statistical relationships among flight variables using rule induction, (2) using a set of rules that cover all features to define the input space of multiple CVAEs, and (3) forming a belief space based on the joint probability density of their collective latent spaces. This results in a series of relatively small matrix multiplications that can be performed in real time, as opposed to large matrix computations in a single CVAE. We demonstrate the application of this approach on the T-2 flight loss-of control experiments, using the architecture and hyperparameter recommendations from the case study. We compare the utilities of an individual CVAE trained on all flight variables and multiple CVAEs defined on subsets of flight variables for detecting qualitative changes in flight. We demonstrate that the use of multiple CVAEs with smaller input spaces permits the CVAE to capture more granular relationships in the latent space, permitting better state space characterization and loss-of-control detection.
Design of experiments↗
A fast algorithm for spectral differentiation
A simple algorithm is presented for matrix multiplication in just over half the number of operations entailed by the conventional algorithm, in cases where the matrix possesses the degree of symmetry widely exhibited by derivative matrices. The algorithm is used to multiply the Chebyshev derivative matrix by a vector. For the larger values of n, the ratio of execution times approached the expected value of 2.
Solomonoff, Alex↗
Durability and Damage Development in Woven Ceramic Matrix Composites
Damage development in woven SiC/SiNC ceramic matrix composites (CMC's) under tensile and cyclic loading both at room and elevated temperatures have been investigated for the exhaust nozzle of high-efficient turbine engines. The ultimate strength, failure strain, proportional limit and modulus data at a temperature range of 23 to 1250 C are generated. The tensile strength of SiC/SiNC woven composites have been observed to increase with increased temperatures up to 1000 C. The stress/strain plot shows a pseudo-yield point at 25 percent of the failure strain (epsilon(sub r)) which indicates damage initiation in the form of matrix cracking. The evolution of damage beyond 0.25 epsilon(sub f), both at room and elevated temperature comprises multiple matrix cracking, interfacial debonding, and fiber pullout. Although the nature of the stress/strain plot shows damage-tolerant behavior under static loading both at room and elevated temperature, the life expectancy of SiC/SiNC composites degrades significantly under cyclic loading at elevated temperature. This is mostly due to the interactions of fatigue damage caused by the mechanically induced plastic strain and the damage developed by the creep strain. The in situ damage evolutions are monitored by acoustic event parameters, ultrasonic C-scan and stiffness degradation. Rate equations for modulus degradation and fatigue life prediction of ceramic matrix composites both at room and elevated temperatures are developed. These rate equations are observed to show reasonable agreement with experimental results.
Haque, A.↗
Understanding Mixed Precision GEMM with MPGemmFI: Insights into Fault Resilience
Emerging deep learning workloads urgently need fast general matrix multiplication (GEMM). Thus, one of the critical features of machine-learning-specific accelerators such as NVIDIA Tensor Cores, AMD Matrix Cores, and Google TPUs is the support of mixed-precision enabled GEMM. For DNN models, lower-precision FP data formats and computation offer acceptable correctness but significant performance, area, and memory footprint improvement. While promising, the mixed-precision computation on error resilience remains unexplored. To this end, we develop a fault injection framework that systematically injects fault into the mixed-precision computation results. We investigate how the faults affect the accuracy of machine learning applications. Based on the characteristics of error resilience, we offer lightweight error detection and correction solutions that significantly improve the overall model accuracy by 75% if the models experience hardware faults. The solutions can be efficiently integrated into the accelerator's pipelines.
Fang, Bo↗
Shifting Between Compute and Memory Bounds: A Compression-Enabled Roofline Model
In the evolving landscape of high-performance computing, especially to fight the end of Moore’s Law and Dennard’s Scaling, the ability to shift between compute-bound and memory-bound states is critical for enhancing adaptability and flexibility to diverse system and domain-specific architectures. Such capability is vital for optimizing performance across distinguished hardware configurations, such as accelerators, memory hierarchies, and cache systems. Despite that ad hoc optimization techniques, such as compressed/approximate computation, have been enabled for compute-/data-intensive computing for improved performance in distinct hardware settings, there lacks an understanding of 1) the rational behind performance improvement; 2) capability of different optimizations; 3) what optimization to respond to specific computational and memory demands. This work proposes a compression-enabled roofline model to facilitate this adaptability with data compression techniques to balance and transform between computational and memory demands. This model enables applications to adjust in response to the specific strengths and limitations of the underlying hardware and system to optimize resource utilization. The effectiveness of this approach is demonstrated with matrix multiplication kernels on different input sizes, with turning on/off various compression techniques, including 1) low-precision floating point; 2) sparse matrix formulation; and 3) compressed arrays with ZFP. By reducing memory transfer volumes and cache misses and increasing data locality and computational intensity through compression, the specific roofline model can transform between compute and memory bounds to align more efficiently with system capabilities. This advancement not only improves overall performance but also maximizes adaptability in diverse computing environments.
Naraparaju, Ramasoumya [University of Washington]↗
Efficient Scalable Contact Network Generation from Population Data
Modeling the contacts among a population is critical to understanding the dynamics of a disease outbreak. Contact networks, where nodes are individuals and edges are contacts among them, are used to represent these complex individual-level interactions. In this work, we are given the daily activity schedules of an urban population that represent the activity location and time of individuals in a population during a single twenty four hour period over multiple days. Using collocation to determine contact between individuals, our goal is to extract hourly contact networks from large-scale activity data. We improve upon the existing adjacency matrix-based method by implementing our custom sparse matrix multiplication algorithm. Starting with a Python implementation, we achieve a 1600x speed up in the computation with a fast custom designed sparse matrix multiplier algorithm implemented in the C++ language. This work is central to future parallel designs of the problem.
97 MATHEMATICS AND COMPUTING↗
A Storage-Efficient WY Representation for Products of Householder Transformations
A product Q=P1 ... P(sub r) of m x m Householder matrices can be written in the form Q = I + WY(sup T), where W and Y are each m x r. This is called the WY representation of Q. It is of interest when implementing Householder techniques in high-performance computing environments that are especially good at matrix-matrix multiplication. In this note a storage-efficient way to implement the WY representation is described. In particular, it is shown how the matrix Q can be expressed in the form Q = I + YTY(sup T). Usually r much less than m and so this 'compact' WY representation requires less storage. When compared with the recent block-reflector strategy the new technique still has a storage advantage and involves a comparable amount of work.
Schreiber, Robert↗
A Fast Algorithm for Computing Zigzag Representatives
Zigzag filtrations of simplicial complexes generalize the usual filtrations by allowing simplex deletions in addition to simplex insertions. The barcodes computed from zigzag filtrations encode the evolution of homological features. Although one can locate a particular feature at any index in the filtration using existing algorithms, the resulting representatives may not be compatible with the zigzag: a representative cycle at one index may not map into a representative cycle at its neighbor. For this, one needs to compute compatible representative cycles along each bar in the barcode. It is known that the barcode for a zigzag filtration with m insertions and deletions can be computed $O(m^ω)$ in time, where $ω < 2.373$ is the matrix multiplication exponent. However, it is not known how to compute the compatible representatives so efficiently. For a non-zigzag filtration, the classical matrix-based algorithm provides representatives in $O(m^3)$ time, which can be improved to $O(m^ω)$. However, no known algorithm for zigzag filtrations computes the representatives with the $O(m^3)$ time bound. We present an $O(m^3 n)$ time algorithm for this problem, where $n ≤ m$ is the size of the largest complex in the filtration.
Persistent homology↗
Proving Program Termination With Matrix Weighted Digraphs
Program termination analysis is an important task in logic and computer science. While determining if a program terminates is known to be undecidable in general, there has been a significant amount of attention given to finding sufficient and computationally practical conditions to prove termination. One such method takes a program and builds from it a matrix weighted digraph. These are directed graphs whose edges are labeled by square matrices with entries in {-1,0,1}, equipped with a nonstandard matrix multiplication. Certain properties of this digraph are known to imply the termination of the related program. In particular, termination of the program can be determined from the weights of the circuits in the digraph. In this talk, the motivation for addressing termination and how matrix weighted digraphs arise will be briefly discussed. The remainder of the talk will describe an efficient method for bounding the weights of a finite set of the circuits in a matrix weighted digraph, which allows termination of the related program to be deduced.
Dutle, Aaron↗