On the use of generalized inverses in function minimization
Minimization algorithms unified derivation, using Hessian matrix inverse
SEARCH · Search NASA
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.
Minimization algorithms unified derivation, using Hessian matrix inverse
A general methodology for damper placement in spaceborne interferometers is introduced.
This paper describes the algorithm used to point the high gain antennae on NASA/JPL's Mars exploration rovers. Each rover's gimballed antenna must track the Earth as it moves across the Martian sky during communication sessions. The pointing algorithm accounts for obstacles to the line-of-sight posed by (1) features on the rover and in the surrounding environment (2) gimbal range limitations, and (3) kinematic singularities in the gimbal mechanism. The algorithm treats all obstacles with a generalized approach that computes the intercept-times to each obstacle. Where possible, the algorithm takes advantage of pairs of joint-space solutions arising from the mechanism design. The algorithm chooses the solution that provides the longest obstruction-free tracking time. Upon encountering an obstacle, the algorithm automatically switches to the other solution if it is not also obstructed. This algorithm has successfully provided obstruction-free pointing for both rovers throughout the mission.
We present two algorithms for accurate beam-frame simulations of the dynamical friction force on a non-relativistic ion moving for a short time in a low-density electron distribution, in the presence of arbitrary external fields. A special-purpose 4th-order predictor-corrector ('Hermite') algorithm, taken from the astrophysical dynamics community, has been generalized to work with charged particles in the presence of a constant magnetic field. An alternative algorithm uses operator splitting techniques to solve binary Coulomb collisions (BCC) in the presence of arbitrary external fields. We discuss the close mathematical relationship between the Hermite and BCC algorithms, and their order of convergence. We discuss the parallel efficiency of the BCC algorithm and use it in the parallel simulation framework VORPAL to study problems in a parameter regime relevant to the electron cooling section for the proposed luminosity upgrade of the Relativistic Heavy Ion Collider. In particular, we simulate the field-free case to show how finite time effects strongly modify the traditional Coulomb logarithm, resulting in a significant reduction of the dynamical friction force as calculated by standard theoretical formulas. We show that diffusive dynamics can be correctly simulated, but that it must be artificially suppressed in order to accurately obtain the friction force. We discuss the proposed use of a helical undulator magnet to focus the electron beam and inhibit electron-ion recombination, showing that this device reduces the friction force.
A digital computer algorithm for desensitized specific optimal system design for general time-invariant nonlinear systems is presented. The algorithm utilizes a technique of partitioning and uncoupling of the original specific closed-loop system state and sensitivity equations into linear, nonlinear, and control subsystems. Results of the application of the algorithm in the design of a Saturn V Launch Vehicle analog attitude control system and an LST image motion compensation digital controller are given.
A parallel algorithm is presented for the solution of the generalized eigenproblem in linear elastic finite element analysis. The algorithm is based on a completely connected parallel architecture in which each processor is allowed to communicate with all other processors. The algorithm is successfully implemented on a tightly coupled MIMD parallel processor. A finite element model is divided into m domains each of which is assumed to process n elements. Each domain is then assigned to a processor or to a logical processor (task) if the number of domains exceeds the number of physical processors. The effect of the number of domains, the number of degrees-of-freedom located along the global fronts, and the dimension of the subspace on the performance of the algorithm is investigated. For a 64-element rectangular plate, speed-ups of 1.86, 3.13, 3.18, and 3.61 are achieved on two, four, six, and eight processors, respectively.
Our modern understanding of cosmological structure formation posits that small matter density fluctuations present in the early Universe, as traced by the cosmic microwave background, grow via gravitational instability to form extended haloes of dark matter. A theoretical understanding of the structure, evolution and formation of dark matter haloes is an essential step towards unravelling the intricate connection between halo and galaxy formation, needed to test our cosmological model against data from upcoming galaxy surveys. Physical understanding of the process of dark matter halo formation is made difficult by the highly non-linear nature of the haloes' evolution. I describe a new approach to gain physical insight into cosmological structure formation based on machine learning. This approach combines the ability of machine learning algorithms to learn non-linear relationships, with techniques that enable us to physically interpret the learnt mapping. I describe applicatio ns of th e method, with the aim of investigating which aspects of the early universe density field impact the later formation of dark matter haloes. First I present a case where the process of halo formation is turned into a binary classification problem; the algorithm predicts whether or not dark matter `particles' in the initial conditions of a simulation will collapse into haloes of a given mass range. Second, I present its generalization to regression, where the algorithm infers the final mass of the halo to which each particle will later belong. I show that the initial tidal shear does not play a significant role compared to the initial density field in establishing final halo masses. Finally, I demonstrate that extending the framework to deep learning algorithms such as convolutional neural networks allows us to explore connections between the early universe and late time haloes beyond those studied by existing analytic approximations of halo collapse.
We present an algorithmic framework for quantum-inspired classical algorithms on close-to-low-rank matrices, generalizing the series of results started by Tang’s breakthrough quantum-inspired algorithm for recommendation systems [STOC’19]. Motivated by quantum linear algebra algorithms and the quantum singular value transformation (SVT) framework of Gilyén et al. [STOC’19], we develop classical algorithms for SVT that run in time independent of input dimension, under suitable quantum-inspired sampling assumptions. Our results give compelling evidence that in the corresponding QRAM data structure input model, quantum SVT does not yield exponential quantum speedups. Since the quantum SVT framework generalizes essentially all known techniques for quantum linear algebra, our results, combined with sampling lemmas from previous work, suffice to generalize all prior results about dequantizing quantum machine learning algorithms. In particular, our classical SVT framework recovers and often improves the dequantization results on recommendation systems, principal component analysis, supervised clustering, support vector machines, low-rank regression, and semidefinite program solving. We also give additional dequantization results on low-rank Hamiltonian simulation and discriminant analysis. Our improvements come from identifying the key feature of the quantum-inspired input model that is at the core of all prior quantum-inspired results: ℓ 2 -norm sampling can approximate matrix products in time independent of their dimension. We reduce all our main results to this fact, making our exposition concise, self-contained, and intuitive.
The Lanczos algorithm for the solution of generalized eigen-problem has been receiving a lot of attention in recent years due to its computational efficiency. The focus of this paper is to develop a Lanczos algorithm which can exploit both the parallel and vector capabilities provided by modern high-performance computers. A partial restoring orthogonality scheme is also developed and incorporated into the basic Lanczos algorithm. The numerical performance in terms of accuracy and efficiency of the proposed parallel-vector Lanczos algorithm is demonstrated by solving for the frequencies and mode shapes of structural problems on multiprocessor supercomputers.
We generalize the binary quantum counting algorithm of Lesovik, Suslov, and Blatter [Phys. Rev. A 82, 012316 (2010)] to higher counting bases. The algorithm makes use of qubits, qutrits, and qudits to count numbers in a base-2, base-3, or base-d representation. In operating the algorithm, the number n<N=d{sup K} is read into a K-qudit register through its interaction with a stream of n particles passing in a nearby wire; this step corresponds to a quantum Fourier transformation from the Hilbert space of particles to the Hilbert space of qudit states. An inverse quantum Fourier transformation provides the number n in the base-d representation; the inverse transformation is fully quantum at the level of individual qudits, while a simpler semiclassical version can be used on the level of qudit registers. Combining registers of qubits, qutrits, and qudits, where d is a prime number, with a simpler single-shot measurement allows us to find the powers of 2, 3, and other primes d in the number n. We show that the counting task naturally leads to the shift operation and an algorithm based on the quantum Fourier transformation. We discuss possible implementations of the algorithm using quantum spin-d systems, d-well systems, and their emulation with spin-1/2 or double-well systems. We establish the analogy between our counting algorithm and the phase estimation algorithm and make use of the latter's performance analysis in stabilizing our scheme. Applications embrace a quantum metrological scheme to measure voltage (an analog to digital converter) and a simple procedure to entangle multiparticle states.
Traditional expert systems, such as diagnostic and training systems, interact with users only through a keyboard and screen, and are usually symbolic in nature. Expert systems that require access to data bases, complex simulations and real-time instrumentation have both symbolic as well as algorithmic computing needs. These needs could both be met using a general purpose workstation running both symbolic and algorithmic code, or separate, specialized computers networked together. The latter approach was chosen to implement TEXSYS, the thermal expert system, developed by NASA Ames Research Center in conjunction with Johnson Space Center to demonstrate the ability of an expert system to autonomously monitor the thermal control system of the space station. TEXSYS has been implemented on a Symbolics workstation, and will be linked to a microVAX computer that will control a thermal test bed. This paper will explore the integration options, and present several possible solutions.
Expert systems that require access to data bases, complex simulations and real time instrumentation have both symbolic as well as algorithmic computing needs. These needs could both be met using a general computing workstation running both symbolic and algorithmic code, or separate, specialized computers networked together. The later approach was chosen to implement TEXSYS, the thermal expert system, developed to demonstrate the ability of an expert system to autonomously control the thermal control system of the space station. TEXSYS has been implemented on a Symbolics workstation, and will be linked to a microVAX computer that will control a thermal test bed. Integration options are explored and several possible solutions are presented.
Expert systems that require access to data bases, complex simulations and real time instrumentation have both symbolic and algorithmic needs. Both of these needs could be met using a general purpose workstation running both symbolic and algorithmic codes, or separate, specialized computers networked together. The later approach was chosen to implement TEXSYS, the thermal expert system, developed by the NASA Ames Research Center in conjunction with the Johnson Space Center to demonstrate the ability of an expert system to autonomously monitor the thermal control system of the space station. TEXSYS has been implemented on a Symbolics workstation, and will be linked to a microVAX computer that will control a thermal test bed. The integration options and several possible solutions are presented.
The Arnoldi (1951) and Lanczos (1950) iterative algorithms are used to obtain a few selected eigenvalues of the generalized eigenvalue problem A(x) = lambdaB(x), where A and B are n x n nonsymmetric banded complex matrices. A shift-and-invert strategy is used to increase the rate of convergence toward the desired eigenvalues, and the two approaches are evaluated in view of their performance in a model problem concerning the linear stability of compressible boundary layers. A general scheme for computing the eigenvalues that lie within a 'box' in the complex plane is presented.
We reveal a close relationship between quantum metrology and the Deutsch-Jozsa algorithm on continuous-variable quantum systems. We develop a general procedure, characterized by two parameters, that unifies parameter estimation and the Deutsch-Jozsa algorithm. Depending on which parameter we keep constant, the procedure implements either the parameter-estimation protocol or the Deutsch-Jozsa algorithm. The parameter-estimation part of the procedure attains the Heisenberg limit and is therefore optimal. Due to the use of approximate normalizable continuous-variable eigenstates, the Deutsch-Jozsa algorithm is probabilistic. The procedure estimates a value of an unknown parameter and solves the Deutsch-Jozsa problem without the use of any entanglement.
This paper introduces HRSSA (Hybrid Rejection-based Stochastic Simulation Algorithm), a new efficient hybrid stochastic simulation algorithm for spatially homogeneous biochemical reaction networks. HRSSA is built on top of RSSA, an exact stochastic simulation algorithm which relies on propensity bounds to select next reaction firings and to reduce the average number of reaction propensity updates needed during the simulation. HRSSA exploits the computational advantage of propensity bounds to manage time-varying transition propensities and to apply dynamic partitioning of reactions, which constitute the two most significant bottlenecks of hybrid simulation. A comprehensive set of simulation benchmarks is provided for evaluating performance and accuracy of HRSSA against other state of the art algorithms.
While Markov chain Monte Carlo (MCMC) methods are frequently used for difficult calculations in a wide range of scientific disciplines, they suffer from a serious limitation: their samples are not independent and identically distributed. Consequently, estimates of expectations are biased if the initial value of the chain is not drawn from the target distribution. Regenerative simulation provides an elegant solution to this problem. In this article, we propose a simple regenerative MCMC algorithm to generate variates for any distribution
The weblog, or blog, has become a popular form of social media, through which authors can write posts, which can in turn generate feedback in the form of user comments. When considered in totality, a collection of blogs can thus be viewed as a sort of informal collection of mass sentiment and opinion. An obvious topic of interest might be to mine this collection to obtain some gauge of public sentiment over the wide variety of topics contained therein. However, the sheer size of the so-called blogosphere, combined with the fact that the subjects of posts can vary over a practically limitless number of topics poses some serious challenges when any meaningful analysis is attempted. Namely, the fact that largely anyone with access to the Internet can author their own blog, raises the serious issue of credibility— should some blogs be considered to be more influential than others, and consequently, when gauging sentiment with respect to a topic, should some blogs be weighted more heavily than others? In addition, as new posts and comments can be made on almost a constant basis, any blog analysis algorithm must be able to handle such updates efficiently. In this paper, we give a formalization of the blog model. We give formal methods of quantifying sentiment and influence with respect to a hierarchy of topics, with the specific aim of facilitating the computation of a per-topic, influence-weighted sentiment measure. Finally, as efficiency is a specific endgoal, we give upper bounds on the time required to update these values with new posts, showing that our analysis and algorithms are scalable.