Downloads & Free Reading Options - Results

The fastest known algorithms for the solution of a large elliptic boundary value problem on a massively parallel hypercube all require O(log(n)) floating point operations and O(log(n)) distance-1 communications, if massively parallel is defined to mean a number of processors proportional to the size n of the problem. The Totally Parallel Multilevel Algorithm (TPMA) that has, as special cases, four of these fast algorithms is described. These four algorithms are Parallel Superconvergent Multigrid (PSMG), Robust Multigrid, the Fast Fourier Transformation (FFT) based Spectral Algorithm, and Parallel Cyclic Reduction. The algorithm TPMA, when described recursively, has four steps: (1) project to a collection of interlaced, coarser problems at the next lower level; (2) apply TPMA, recursively, to each of these lower level problems, solving directly at the lowest level; (3) interpolate these approximate solutions to the finer grid, and to verage them to form an approximate solution on this grid; and (4) refine this approximate solution with a defect-correction step, using a local approximate inverse. Choice of the projection operator (P), the interpolation operator (Q), and the smoother (S) determines the class of problems on which TPMA is most effective. There are special cases in which the first three steps produce an exact solution, and the smoother is not needed (e.g., constant coefficient operators).

Solving large, sparse, linear systems of equations is one of the fundamental problems in large scale scientific and engineering computation. A model of a general class of asynchronous, iterative solution methods for linear systems is developed. In the model, the system is solved by creating several cooperating tasks that each compute a portion of the solution vector. This model is then analyzed to determine the expected intertask data transfer and task computational complexity as functions of the number of tasks. Based on the analysis, recommendations for task partitioning are made. These recommendations are a function of the sparseness of the linear system, its structure (i.e., randomly sparse or banded), and dimension.

Call a spectrum of Hamiltonian sparse if each eigenvalue can be quickly restored with accuracy $\epsilon$ from its rough approximation in within $\epsilon_1$ by means of some classical algorithm. It is shown how a behavior of system with sparse spectrum up to time $T=\frac{1-\rho}{14\epsilon}$ can be predicted with fidelity $\rho$ on quantum computer in time $t=\frac{4}{(1-\rho)\epsilon_1}$ plus the time of classical algorithm. The quantum knowledge of Hamiltonian $H$ eigenvalues is considered as a wizard Hamiltonian $W_H$ which action on any eigenvector of $H$ gives the corresponding eigenvalue. Speedup of evolution for systems with sparse spectrum is possible because for such systems wizard Hamiltonians can be quickly simulated on a quantum computer. This simulation, generalizing Shor trick, is a part of presented algorithm. In general case the action of wizard Hamiltonian cannot be simulated in time smaller than the dimension of main space which is exponential of the size of quantum system. For an arbitrary system (even for classical) its behavior cannot be predicted on quantum computer even for one step ahead. This method can be used also for restoration of a state of an arbitrary primary system in time instant $-T$ in the past with the same fidelity which requires the same time.

We consider linear dynamical systems of ordinary differential equations or differential algebraic equations. Physical parameters are substituted by random variables for an uncertainty quantification. We expand the state variables as well as a quantity of interest into an orthogonal system of basis functions, which depend on the random variables. For example, polynomial chaos expansions are applicable. The stochastic Galerkin method yields a larger linear dynamical system, whose solution approximates the unknown coefficients in the expansions. The Hardy norms of the transfer function provide information about the input-output behaviour of the Galerkin system. We investigate two approaches to construct a sparse representation of the quantity of interest, where just a low number of coefficients is non-zero. Firstly, a standard basis is reduced by the omission of basis functions, whose accompanying Hardy norms are relatively small. Secondly, a projection-based model order reduction is applied to the Galerkin system and allows for the definition of new basis functions as a sparse representation. In both cases, we prove error bounds on the sparse approximation with respect to Hardy norms. Numerical experiments are demonstrated for a test example modelling a linear electric circuit.

Polynomial chaos expansions (PCE) have proven efficiency in a number of fields for propagating parametric uncertainties through computational models of complex systems, namely structural and fluid mechanics, chemical reactions and electromagnetism, etc. For problems involving oscillatory, time-dependent output quantities of interest, it is well-known that reasonable accuracy of PCE-based approaches is difficult to reach in the long term. In this paper, we propose a fully non-intrusive approach based on stochastic time warping to address this issue: each realization (trajectory) of the model response is first rescaled to its own time scale so as to put all sampled trajectories in phase in a common virtual time line. Principal component analysis is introduced to compress the information contained in these transformed trajectories and sparse PCE representations using least angle regression are finally used to approximate the components. The approach shows remarkably small prediction error for particular trajectories as well as for second-order statistics of the latter. It is illustrated on different benchmark problems well known in the literature on time-dependent PCE problems, ranging from rigid body dynamics, chemical reactions to forced oscillations of a non linear system.

We assume the direct sum o for the signal subspace. As a result of post- measurement, a number of operational contexts presuppose the a priori knowledge of the LB -dimensional "interfering" subspace and the goal is to estimate the LA am- plitudes corresponding to subspace

The stripe structures of stiffness matrices resulting from irregular domains covered by regular grids; are analysed. It is proved that the non-zero elements in these matrices may be covered by very few stripes, and that these stripes may be non-overlapping, if the nodes of the grids are numbered appropriately. The exact number of stripes, which is independent of the size of the problem, is derived for different types of grids, and different numbering schemes. The stripe structure of some irregular grids are also examined. Keywords: Parallel processing; partial differential equations.

A resistor-network picture of transitions is appropriate for the study of energy absorption by weakly chaotic or weakly interacting driven systems. Such "sparse" systems reach a novel non-equilibrium steady state (NESS) once coupled to a bath. In the stochastic case there is an analogy to the physics of percolating glassy systems, and an extension of the fluctuation-dissipation phenomenology is proposed. In the mesoscopic case the quantum NESS might differ enormously from the stochastic NESS, with saturation temperature determined by the sparsity. A toy model where the sparsity of the system is modeled using a log-normal random ensemble is analyzed.

Nowadays sparse systems of equations occur frequently in science and engineering. In this contribution we deal with sparse systems common in cryptanalysis. Given a cipher system, one converts it into a system of sparse equations, and then the system is solved to retrieve either a key or a plaintext. Raddum and Semaev proposed new methods for solving such sparse systems. It turns out that a combinatorial MaxMinMax problem provides bounds on the average computational complexity of sparse systems. In this paper we initiate a study of a linear algebra variation of this MaxMinMax problem.

In this paper, we introduce a new detection algorithm for large-scale wireless systems, referred to as post sparse error detection (PSED) algorithm, that employs a sparse error recovery algorithm to refine the estimate of a symbol vector obtained by the conventional linear detector. The PSED algorithm operates in two steps: 1) sparse transformation converting the original non-sparse system into the sparse system whose input is an error vector caused by the symbol slicing and 2) estimation of the error vector using the sparse recovery algorithm. From the asymptotic mean square error (MSE) analysis and empirical simulations performed on large-scale systems, we show that the PSED algorithm brings significant performance gain over classical linear detectors while imposing relatively small computational overhead.

While many of the architectural details of future exascale-class high performance computer systems are still a matter of intense research, there appears to be a general consensus that they will be strongly heterogeneous, featuring "standard" as well as "accelerated" resources. Today, such resources are available as multicore processors, graphics processing units (GPUs), and other accelerators such as the Intel Xeon Phi. Any software infrastructure that claims usefulness for such environments must be able to meet their inherent challenges: massive multi-level parallelism, topology, asynchronicity, and abstraction. The "General, Hybrid, and Optimized Sparse Toolkit" (GHOST) is a collection of building blocks that targets algorithms dealing with sparse matrix representations on current and future large-scale systems. It implements the "MPI+X" paradigm, has a pure C interface, and provides hybrid-parallel numerical kernels, intelligent resource management, and truly heterogeneous parallelism for multicore CPUs, Nvidia GPUs, and the Intel Xeon Phi. We describe the details of its design with respect to the challenges posed by modern heterogeneous supercomputers and recent algorithmic developments. Implementation details which are indispensable for achieving high efficiency are pointed out and their necessity is justified by performance measurements or predictions based on performance models. The library code and several applications are available as open source. We also provide instructions on how to make use of GHOST in existing software packages, together with a case study which demonstrates the applicability and performance of GHOST as a component within a larger software stack.

We consider families of sparse Laurent polynomials f_1,...,f_n with a finite set of common zeroes Z_f in the complex algebraic n-torus. The global residue assigns to every Laurent polynomial g the sum of its Grothendieck residues over the set Z_f. We present a new symbolic algorithm for computing the global residue as a rational function of the coefficients of the f_i when the Newton polytopes of the f_i are full-dimensional. Our results have consequences in sparse polynomial interpolation and lattice point enumeration in Minkowski sums of polytopes.

Consider a system F of n polynomial equations in n unknowns, over an algebraically closed field of arbitrary characteristic. We present a fast method to find a point in every irreducible component of the zero set Z of F. Our techniques allow us to sharpen and lower prior complexity bounds for this problem by fully taking into account the monomial term structure. As a corollary of our development we also obtain new explicit formulae for the exact number of isolated roots of F and the intersection multiplicity of the positive-dimensional part of Z. Finally, we present a combinatorial construction of non-degenerate polynomial systems, with specified monomial term structure and maximally many isolated roots, which may be of independent interest.

Vector orthogonal frequency division multiplexing (V-OFDM) is a general system that builds a bridge between OFDM and single-carrier frequency domain equalization in terms of intersymbol interference and receiver complexity. In this paper, we investigate the sparse multipath channel estimation and decoding for broadband V-OFDM systems. Unlike the non-sparse channel estimation, sparse channel estimation only needs to recover the nonzero taps with reduced complexity. Consider the pilot signals are transmitted through a sparse channel that has only a few nonzero taps with and without additive white Gaussian noise, respectively. The exactly and approximately sparse inverse fast Fourier transform (SIFFT) can be employed for these two cases. The SIFFT-based algorithm recovers the nonzero channel coefficients and their corresponding coordinates directly, which is significant to the proposed partial intersection sphere (PIS) decoding approach. Unlike the maximum likelihood (ML) decoding that enumerates symbol constellation and estimates the transmitted symbols with the minimum distance, the PIS decoding first generates the set of possible transmitted symbols and then chooses the transmitted symbols only from this set with the minimum distance. The diversity order of the PIS decoding is determined by not only the number of nonzero taps, but also the coordinates of nonzero taps, and the bit error rate (BER) is also influenced by vector block size to some extent but roughly independent of the maximum time delay. Simulation results indicate that by choosing appropriate sphere radius, the BER performance of the PIS decoding outperforms the conventional zero-forcing decoding and minimum mean square error decoding, and approximates to the ML decoding with the increase of signal-to-noise ratio, but reduces the computational complexity significantly.

We propose an iterative channel estimation algorithm based on the Least Square Estimation (LSE) and Sparse Message Passing (SMP) algorithm for the Millimeter Wave (mmWave) MIMO systems. The channel coefficients of the mmWave MIMO are approximately modeled as a Bernoulli-Gaussian distribution since there are relatively fewer paths in the mmWave channel, i.e., the channel matrix is sparse and only has a few non-zero entries. By leveraging the advantage of sparseness, we proposed an algorithm that iteratively detects the exact location and value of non-zero entries of the sparse channel matrix. The SMP is used to detect the exact location of non-zero entries of the channel matrix, while the LSE is used for estimating its value at each iteration. We also analyze the Cramer-Rao Lower Bound (CLRB), and show that the proposed algorithm is a minimum variance unbiased estimator. Furthermore, we employ the Gaussian approximation for message densities under density evolution to simplify the analysis of the algorithm, which provides a simple method to predict the performance of the proposed algorithm. Numerical experiments show that the proposed algorithm has much better performance than the existing sparse estimators, especially when the channel is sparse. In addition, our proposed algorithm converges to the CRLB of the genie-aided estimation of sparse channels in just 5 turbo iterations.

Consider a regular triangulation of the convex-hull $P$ of a set $\mathcal A$ of $n$ points in $\mathbb R^d$, and a real matrix $C$ of size $d \times n$. A version of Viro's method allows to construct from these data an unmixed polynomial system with support $\mathcal A$ and coefficient matrix $C$ whose number of positive solutions is bounded from below by the number of $d$-simplices which are positively decorated by $C$. We show that all the $d$-simplices of a triangulation can be positively decorated if and only if the triangulation is balanced, which in turn is equivalent to the fact that its dual graph is bipartite. This allows us to identify, among classical families, monomial supports which admit maximally positive systems, i.e. systems all toric complex solutions of which are real and positive. These families give some evidence in favor of a conjecture due to Bihan. We also use this technique in order to construct fewnomial systems with many positive solutions. This is done by considering a simplicial complex with bipartite dual graph included in a regular triangulation of the cyclic polytope.

We define sparse saturated fusion systems and show that, for odd primes, sparse systems are constrained. This simplifies the proof of the Glauberman-Thompson p-nilpotency theorem for fusion systems and a related theorem of Stellmacher. We then define a more restrictive class of saturated fusion systems, called extremely sparse, that are constrained for all primes.

We demonstrate the synthesis of sparse sampling and machine learning to characterize and model complex, nonlinear dynamical systems over a range of bifurcation parameters. First, we construct modal libraries using the classical proper orthogonal decomposition to uncover dominant low-rank coherent structures. Here, nonlinear libraries are also constructed in order to take advantage of the discrete empirical interpolation method and projection that allows for the approximation of nonlinear terms in a low-dimensional way. The selected sampling points are shown to be nearly optimal sensing locations for characterizing the underlying dynamics, stability, and bifurcations of complex systems. The use of empirical interpolation points and sparse representation facilitate a family of local reduced-order models for each physical regime, rather than a higher-order global model, which has the benefit of physical interpretability of energy transfer between coherent structures. In particular, the discrete interpolation points and nonlinear modal libraries are used for sparse representation to classify the dynamic bifurcation regime in the complex Ginzburg-Landau equation. It is shown that nonlinear point measurements are more effective than linear measurements when sensor noise is present.

We consider the problem of computing an approximate banded solution of the continuous-time Lyapunov equation $\underline{A}\underline{X}+\underline{X}\underline{A}^{T}=\underline{P}$, where the coefficient matrices $\underline{A}$ and $\underline{P}$ are large, symmetric banded matrices. The (sparsity) pattern of $\underline{A}$ describes the interconnection structure of a large-scale interconnected system. Recently, it has been shown that the entries of the solution $\underline{X}$ are spatially localized or decaying away from a banded pattern. We show that the decay of the entries of $\underline{X}$ is faster if the condition number of $\underline{A}$ is smaller. By exploiting the decay of entries of $\underline{X}$, we develop two computationally efficient methods for approximating $\underline{X}$ by a banded matrix. For a well-conditioned and sparse banded $\underline{A}$, the computational and memory complexities of the methods scale linearly with the state dimension. We perform extensive numerical experiments that confirm this, and that demonstrate the effectiveness of the developed methods. The methods proposed in this paper can be generalized to (sparsity) patterns of $\underline{A}$ and $\underline{P}$ that are more general than banded matrices. The results of this paper open the possibility for developing computationally efficient methods for approximating the solution of the large-scale Riccati equation by a sparse matrix.

We propose a method for calculating the Franz-Parisi potential for spin glass models on sparse random graphs using the replica method under the replica symmetric ansatz. The resulting self-consistent equations have the solution with the characteristic structure of multi-body overlaps, and the self-consistent equations under this solution are equivalent to the one-step replica symmetry breaking (1RSB) cavity equation with Parisi parameter $x=1$. This method is useful for the evaluation of transition temperatures of the $p$-spin model on regular random graphs under a uniform magnetic field.

We investigate the SPAI and PSAI preconditioning procedures and shed light on two important features of them: (i) For the large linear system $Ax=b$ with $A$ irregular sparse, i.e., with $A$ having $s$ relatively dense columns, SPAI may be very costly to implement, and the resulting sparse approximate inverses may be ineffective for preconditioning. PSAI can be effective for preconditioning but may require excessive storage and be unacceptably time consuming; (ii) the situation is improved drastically when $A$ is regular sparse, that is, all of its columns are sparse. In this case, both SPAI and PSAI are efficient. Moreover, SPAI and, especially, PSAI are more likely to construct effective preconditioners. Motivated by these features, we propose an approach to making SPAI and PSAI more practical for $Ax=b$ with $A$ irregular sparse. We first split $A$ into a regular sparse $\tilde A$ and a matrix of low rank $s$. Then exploiting the Sherman--Morrison--Woodbury formula, we transform $Ax=b$ into $s+1$ new linear systems with the same coefficient matrix $\tilde A$, use SPAI and PSAI to compute sparse approximate inverses of $\tilde A$ efficiently and apply Krylov iterative methods to solve the preconditioned linear systems. Theoretically, we consider the non-singularity and conditioning of $\tilde A$ obtained from some important classes of matrices. We show how to recover an approximate solution of $Ax=b$ from those of the $s+1$ new systems and how to design reliable stopping criteria for the $s+1$ systems to guarantee that the approximate solution of $Ax=b$ satisfies a desired accuracy. Given the fact that irregular sparse linear systems are common in applications, this approach widely extends the practicability of SPAI and PSAI. Numerical results demonstrate the considerable superiority of our approach to the direct application of SPAI and PSAI to $Ax=b$.

For several computational procedures such as finding radicals and Noether normalizations, it is important to choose as sparse as possible a system of parameters in a polynomial ideal or modulo a polynomial ideal. We describe new strategies for these tasks, thus providing solutions to problems (1) and (2) posed in [Eisenbud-Huneke-Vasconcelos 1992].

This article is from EURASIP Journal on Bioinformatics and Systems Biology , volume 2014 . Abstract Parameter estimation in dynamic systems finds applications in various disciplines, including system biology. The well-known expectation-maximization (EM) algorithm is a popular method and has been widely used to solve system identification and parameter estimation problems. However, the conventional EM algorithm cannot exploit the sparsity. On the other hand, in gene regulatory network inference problems, the parameters to be estimated often exhibit sparse structure. In this paper, a regularized expectation-maximization (rEM) algorithm for sparse parameter estimation in nonlinear dynamic systems is proposed that is based on the maximum a posteriori (MAP) estimation and can incorporate the sparse prior. The expectation step involves the forward Gaussian approximation filtering and the backward Gaussian approximation smoothing. The maximization step employs a re-weighted iterative thresholding method. The proposed algorithm is then applied to gene regulatory network inference. Results based on both synthetic and real data show the effectiveness of the proposed algorithm.

Recently, there has been a significant interest in developing cooperative sensing systems for certain types of wireless applications. In such systems, a group of sensing nodes periodically collect measurements about the signals being observed in the given geographical region and transmit these measurements to a central node, which in turn processes this information to recover the signals. For example, in cognitive radio networks, the signals of interest are those generated by the primary transmitters and the sensing nodes are the secondary users. In such networks, it is critically important to be able to reliably determine the presence or absence of primary transmitters in order to avoid causing interference. The standard approach to transmit these measurements from sensor the nodes to the fusion center has been to use orthogonal channels. Such an approach quickly places a burden on the control-channel-capacity of the network that would scale linearly in the number of cooperating sensing nodes. In this paper, we show that as long as one condition is satisfied: the dynamics of the observed signals are sparse, i.e., the observed signals do not change their values very rapidly in relation to the time-scale at which the measurements are collected, we can significantly reduce the control bandwidth of the system while achieving the full (linear) bandwidth performance.

Bayesian estimation strategies represent the most fundamental formulation of the state estimation problem available, and apply readily to nonlinear systems with non-Gaussian uncertainties. The present paper introduces a novel method for implementing grid-based Bayesian estimation which largely sidesteps the severe computational expense that has prevented the widespread use of such methods. The method represents the evolution of the probability density function (PDF) in phase space, $p_{\x}(\x',t)$, discretized on a fixed Cartesian grid over {\it all} of phase space, and consists of two main steps: (i) Between measurement times, $p_{\x}(\x',t)$ is evolved via numerical discretization of the Kolmogorov forward equation, using a Godunov method with second-order corner transport upwind correction and a total variation diminishing flux limiter; (ii) at measurement times, $p_{\x}(\x',t)$ is updated via Bayes' theorem. Computational economy is achieved by exploiting the localised nature of $p_{\x}(\x',t)$. An ordered list of cells with non-negligible probability, as well as their immediate neighbours, is created and updated, and the PDF evolution is tracked {\it only} on these active cells. %The grid-based discretization of $p_{\x}(\x',t)$ in this approach avoids the requirement for resampling associated with particle-based representations of the PDF.

The SPAI algorithm, a sparse approximate inverse preconditioning technique for large sparse linear systems, proposed by Grote and Huckle [SIAM J. Sci. Comput., 18 (1997), pp.~838--853.], is based on the F-norm minimization and computes a sparse approximate inverse $M$ of a large sparse matrix $A$ adaptively. However, SPAI may be costly to seek the most profitable indices at each loop and $M$ may be ineffective for preconditioning. In this paper, we propose a residual based sparse approximate inverse preconditioning procedure (RSAI), which, unlike SPAI, is based on only the {\em dominant} rather than all information on the current residual and augments sparsity patterns adaptively during the loops. RSAI is less costly to seek indices and is more effective to capture a good approximate sparsity pattern of $A^{-1}$ than SPAI. To control the sparsity of $M$ and reduce computational cost, we develop a practical RSAI($tol$) algorithm that drops small nonzero entries adaptively during the process. Numerical experiments are reported to demonstrate that RSAI($tol$) is at least competitive with SPAI and can be considerably more efficient and effective than SPAI. They also indicate that RSAI($tol$) is comparable to the PSAI($tol$) algorithm proposed by one of the authors in 2009.

Cooperative systems are systems in which the forces among agents are non-repulsive. The free evolution of such systems can tend to the formation of patterns, such as consensus or clustering, depending on the properties and intensity of the interaction forces between agents. The kinetic cooperative systems are obtained as the mean field limits of these systems when the number of agents goes to infinity. These limit dynamics are described by transport partial differential equations involving non-local terms. In this article, we design a simple and robust control strategy steering any kinetic cooperative system to approximate alignment. The computation of the control at each instant will only require knowledge of the size of the support of the crowd in the phase space and of the Lipschitz constant of the interaction forces. Besides, the control we apply to our system is sparse, in the sense that it acts only on a small portion of the total population at each time. It also presents the features of being obtained through a constructive procedure and to be independent on the number of agents, making it convenient for applications.

We introduce low complexity machine learning based approach for mitigating nonlinear impairments in optical fiber communications systems. The immense intricacy of the problem calls for the development of "smart" methodology, simplifying the analysis without losing the key features that are important for recovery of transmitted data. The proposed sparse identification method for optical systems (SINO) allows to determine the minimal (optimal) number of degrees of freedom required for adaptive mitigation of detrimental nonlinear effects. We demonstrate successful application of the SINO method both for standard fiber communication links and for few-mode spatial-division-multiplexing systems.

We develop an algorithm for model selection which allows for the consideration of a combinatorially large number of candidate models governing a dynamical system. The innovation circumvents a disadvantage of standard model selection which typically limits the number candidate models considered due to the intractability of computing information criteria. Using a recently developed sparse identification of nonlinear dynamics algorithm, the sub-selection of candidate models near the Pareto frontier allows for a tractable computation of AIC (Akaike information criteria) or BIC (Bayes information criteria) scores for the remaining candidate models. The information criteria hierarchically ranks the most informative models, enabling the automatic and principled selection of the model with the strongest support in relation to the time series data. Specifically, we show that AIC scores place each candidate model in the {\em strong support}, {\em weak support} or {\em no support} category. The method correctly identifies several canonical dynamical systems, including an SEIR (susceptible-exposed-infectious-recovered) disease model and the Lorenz equations, giving the correct dynamical system as the only candidate model with strong support.

The ability of computers to solve hitherto intractable problems and simulate complex processes using mathematical models makes them an indispensable part of modern science and engineering. Computer simulations of large-scale realistic applications usually require solving a set of non-linear partial differential equations (PDES) over a finite region. For example, one thrust area in the DOE Grand Challenge projects is to design future accelerators such as the SpaHation Neutron Source (SNS). Our colleagues at SLAC need to model complex RFQ cavities with large aspect ratios. Unstructured grids are currently used to resolve the small features in a large computational domain; dynamic mesh adaptation will be added in the future for additional efficiency. The PDEs for electromagnetics are discretized by the FEM method, which leads to a generalized eigenvalue problem Kx = AMx, where K and M are the stiffness and mass matrices, and are very sparse. In a typical cavity model, the number of degrees of freedom is about one million. For such large eigenproblems, direct solution techniques quickly reach the memory limits. Instead, the most widely-used methods are Krylov subspace methods, such as Lanczos or Jacobi-Davidson. In all the Krylov-based algorithms, sparse matrix-vector multiplication (SPMV) must be performed repeatedly. Therefore, the efficiency of SPMV usually determines the eigensolver speed. SPMV is also one of the most heavily used kernels in large-scale numerical simulations.

We exhibit a probabilistic symbolic algorithm for solving zero-dimensional sparse systems. Our algorithm combines a symbolic homotopy procedure, based on a flat deformation of a certain morphism of affine varieties, with the polyhedral deformation of Huber and Sturmfels. The complexity of our algorithm is quadratic in the size of the combinatorial structure of the input system. This size is mainly represented by the mixed volume of Newton polytopes of the input polynomials and an arithmetic analogue of the mixed volume associated to the deformations under consideration.

Spreading according to simple rules (e.g. of fire or diseases), and shortest-path distances are studied on d-dimensional systems with a small density p per site of long-range connections (``Small-World'' lattices). The volume V(t) covered by the spreading quantity on an infinite system is exactly calculated in all dimensions. We find that V(t) grows initially as t^d/d for t < < t^* = (2p \Gamma_d (d-1)!)^{-1/d} and later exponentially for $t>>t^*$, generalizing a previous result in one dimension. Using the properties of V(t), the average shortest-path distance \ell(r) can be calculated as a function of Euclidean distance r. It is found that \ell(r) = r for rr_c. The characteristic length r_c, which governs the behavior of shortest-path lengths, diverges with system size for all p>0. Therefore the mean separation s \sim p^{-1/d} between shortcut-ends is not a relevant internal length-scale for shortest-path lengths. We notice however that the globally averaged shortest-path length, divided by L, is a function of L/s only.

The ongoing hardware evolution exhibits an escalation in the number, as well as in the heterogeneity, of computing resources. The pressure to maintain reasonable levels of performance and portability forces application developers to leave the traditional programming paradigms and explore alternative solutions. PaStiX is a parallel sparse direct solver, based on a dynamic scheduler for modern hierarchical manycore architectures. In this paper, we study the benefits and limits of replacing the highly specialized internal scheduler of the PaStiX solver with two generic runtime systems: PaRSEC and StarPU. The tasks graph of the factorization step is made available to the two runtimes, providing them the opportunity to process and optimize its traversal in order to maximize the algorithm efficiency for the targeted hardware platform. A comparative study of the performance of the PaStiX solver on top of its native internal scheduler, PaRSEC, and StarPU frameworks, on different execution environments, is performed. The analysis highlights that these generic task-based runtimes achieve comparable results to the application-optimized embedded scheduler on homogeneous platforms. Furthermore, they are able to significantly speed up the solver on heterogeneous environments by taking advantage of the accelerators while hiding the complexity of their efficient manipulation from the programmer.

The analysis of the response to driving in the case of weakly chaotic or weakly interacting systems should go beyond linear response theory. Due to the "sparsity" of the perturbation matrix, a resistor network picture of transitions between energy levels is essential. The Kubo formula is modified, replacing the "algebraic" average over the squared matrix elements by a "resistor network" average. Consequently the response becomes semi-linear rather than linear. Some novel results have been obtained in the context of two prototype problems: the heating rate of particles in Billiards with vibrating walls; and the Ohmic Joule conductance of mesoscopic rings driven by electromotive force. Respectively, the obtained results are contrasted with the "Wall formula" and the "Drude formula".

In our companion work \cite{Stojnicl1RegPosasymldp} we revisited random under-determined linear systems with sparse solutions. The main emphasis was on the performance analysis of the $\ell_1$ heuristic in the so-called asymptotic regime, i.e. in the regime where the systems' dimensions are large. Through an earlier sequence of work \cite{DonohoPol,DonohoUnsigned,StojnicCSetam09,StojnicUpper10}, it is now well known that in such a regime the $\ell_1$ exhibits the so-called \emph{phase transition} (PT) phenomenon. \cite{Stojnicl1RegPosasymldp} then went much further and established the so-called \emph{large deviations principle} (LDP) type of behavior that characterizes not only the breaking points of the $\ell_1$'s success but also the behavior in the entire so-called \emph{transition zone} around these points. Both of these concepts, the PTs and the LDPs, are in fact defined so that one can use them to characterize the asymptotic behavior. In this paper we complement the results of \cite{Stojnicl1RegPosasymldp} by providing an exact detailed analysis in the non-asymptotic regime. Of course, not only are the non-asymptotic results complementing those from \cite{Stojnicl1RegPosasymldp}, they actually are the ones that ultimately fully characterize the $\ell_1$'s behavior in the most general sense. We introduce several novel high-dimensional geometry type of strategies that enable us to eventually determine the $\ell_1$'s behavior.

The implementation of Pan and Reif's Parallel Nested Dissection (PND) algorithm on mesh connected parallel computers is described. This is the first known algorithm that allows very large, sparse linear systems of equations to be solved efficiently in polylog time using a small number of processors. How the processor bound of PND can be matched to the number of processors available on a given parallel computer by slowing down the algorithm by constant factors is described. Also, for the important class of problems where G(A) is a grid graph, a unique memory mapping that reduces the inter-processor communication requirements of PND to those that can be executed on mesh connected parallel machines is detailed. A description of an implementation on the Goodyear Massively Parallel Processor (MPP), located at Goddard is given. Also, a detailed discussion of data mappings and performance issues is given.

In this paper, we propose a novel channel estimation algorithm based on the Least Square Estimation (LSE) and Sparse Message Passing algorithm (SMP), which is of special interest for Millimeter Wave (mmWave) systems, since this algorithm can leverage the inherent sparseness of the mmWave channel. Our proposed algorithm will iteratively detect exact the location and the value of non-zero entries of sparse channel vector without its prior knowledge of distribution. The SMP is used to detect exact the location of non-zero entries of the channel vector, while the LSE is used for estimating its value at each iteration. Then, the analysis of the Cramer-Rao Lower Bound (CRLB) of our proposed algorithm is given. Numerical experiments show that our proposed algorithm has much better performance than the existing sparse estimators (e.g. LASSO), especially when mmWave systems have massive antennas at both the transmitters and receivers. In addition, we also find that our proposed algorithm converges to the CRLB of the genie-aided estimation of sparse channels in just a few turbo iterations.

This paper presents an algorithm the combined Schubert/secant/finite difference algorithm, for solving sparse nonlinear systems of equations. This algorithm is based on dividing the columns of the Jacobian into two parts, and using different algorithms on each part. This algorithm incorporates advantages of both algorithms by exploiting some special structure of the Jacobian to obtain a good approximation to the Jacobian by using as little effort as possible. Kantorovich-type analysis and a locally q-superlinear convergence result for this algorithm are given.

Initial ranging constitutes a part of the synchronization procedure employed by the wireless communication standards. This allows the base station (BS) to detect the subscriber stations (SS) that are willing to commence communication. In addition, the ranging process allows the BS to estimate the uplink channel parameters of these SSs. Accurate estimation of these parameters are crucial as they ensure that the uplink signals from all the SSs arrive at the BS synchronously and approximately at the same power level. However, this detection and estimation problem turns out to be very challenging when multiple users initiate the ranging procedure at the same time. We address this issue by exploiting the underlying sparsity of the estimation problem. We propose a fast sparse signal recovery approach to improve the ranging performance in multi-user environment. Compared to the standard correlation based techniques, our method shows a clear improvement in ranging code detection, timing offset and channel power estimation. Although this method has been developed around the WiMAX standard, the underlying principles apply to other OFDM based standards as well.

We propose a new algorithm to solve sparse linear systems of equations over the integers. This algorithm is based on a $p$-adic lifting technique combined with the use of block matrices with structured blocks. It achieves a sub-cubic complexity in terms of machine operations subject to a conjecture on the effectiveness of certain sparse projections. A LinBox-based implementation of this algorithm is demonstrated, and emphasizes the practical benefits of this new method over the previous state of the art.

In solving a linear system with iterative methods, one is usually confronted with the dilemma of having to choose between cheap, inefficient iterates over sparse search directions (e.g., coordinate descent), or expensive iterates in well-chosen search directions (e.g., conjugate gradients). In this paper, we propose to interpolate between these two extremes, and show how to perform cheap iterations along non-sparse search directions, provided that these directions can be extracted from a new kind of sparse factorization. For example, if the search directions are the columns of a hierarchical matrix, then the cost of each iteration is typically logarithmic in the number of variables. Using some graph-theoretical results on low-stretch spanning trees, we deduce as a special case a nearly-linear time algorithm to approximate the minimal norm solution of a linear system $Bx= b$ where $B$ is the incidence matrix of a graph. We thereby can connect our results to recently proposed nearly-linear time solvers for Laplacian systems, which emerge here as a particular application of our sparse matrix factorization.

In this paper we show that inverses of well-conditioned, finite-time Gramians and impulse response matrices of large-scale interconnected systems described by sparse state-space models, can be approximated by sparse matrices. The approximation methodology established in this paper opens the door to the development of novel methods for distributed estimation, identification and control of large-scale interconnected systems. The novel estimators (controllers) compute local estimates (control actions) simply as linear combinations of inputs and outputs (states) of local subsystems. The size of these local data sets essentially depends on the condition number of the finite-time observability (controllability) Gramian. Furthermore, the developed theory shows that the sparsity patterns of the system matrices of the distributed estimators (controllers) are primarily determined by the sparsity patterns of state-space matrices of large-scale systems. The computational and memory complexity of the approximation algorithms are $O(N)$, where $N$ is the number of local subsystems of the interconnected system. Consequently, the proposed approximation methodology is computationally feasible for interconnected systems with an extremely large number of local subsystems.

We consider the problem of finding a low discrepancy coloring for sparse set systems where each element lies in at most $t$ sets. We give an algorithm that finds a coloring with discrepancy $O((t \log n \log s)^{1/2})$ where $s$ is the maximum cardinality of a set. This improves upon the previous constructive bound of $O(t^{1/2} \log n)$ based on algorithmic variants of the partial coloring method, and for small $s$ (e.g.$s=\textrm{poly}(t)$) comes close to the non-constructive $O((t \log n)^{1/2})$ bound due to Banaszczyk. Previously, no algorithmic results better than $O(t^{1/2}\log n)$ were known even for $s = O(t^2)$. Our method is quite robust and we give several refinements and extensions. For example, the coloring we obtain satisfies the stronger size-sensitive property that each set $S$ in the set system incurs an $O((t \log n \log |S|)^{1/2})$ discrepancy. Another variant can be used to essentially match Banaszczyk's bound for a wide class of instances even where $s$ is arbitrarily large. Finally, these results also extend directly to the more general Koml\'{o}s setting.

The problem of fault diagnosis in multiprocessor systems is considered under a uniformly probabilistic model in which processors are faulty with probability p. This work focuses on minimizing the number of tests that must be conducted in order to correctly diagnose the state of every processor in the system with high probability. A diagnosis algorithm that can correctly diagnose the state of every processor with probability approaching one in a class of systems performing slightly greater than a linear number of tests is presented. A nearly matching lower bound on the number of tests required to achieve correct diagnosis in arbitrary systems is also proven. The number of tests required under this probabilistic model is shown to be significantly less than under a bounded-size fault set model. Because the number of tests that must be conducted is a measure of the diagnosis overhead, these results represent a dramatic improvement in the performance of system-level diagnosis technique.

We present a new probabilistic symbolic algorithm that, given a variety defined in an n-dimensional affine space by a generic sparse system with fixed supports, computes the Zariski closure of its projection to an l-dimensional coordinate affine space with l < n. The complexity of the algorithm depends polynomially on combinatorial invariants associated to the supports.

This technical note considers the identification of nonlinear discrete-time systems with additive process noise but without measurement noise. In particular, we propose a method and its associated algorithm to identify the system nonlinear functional forms and their associated parameters from a limited number of time-series data points. For this, we cast this identification problem as a sparse linear regression problem and take a Bayesian viewpoint to solve it. As such, this approach typically leads to nonconvex optimisations. We propose a convexification procedure relying on an efficient iterative re-weighted $\ell_1$-minimisation algorithm that uses general sparsity inducing priors on the parameters of the system and marginal likelihood maximisation. Using this approach, we also show how convex constraints on the parameters can be easily added to our proposed iterative re-weighted $\ell_1$-minimisation algorithm. In the supplementary material \cite{appendix}, we illustrate the effectiveness of the proposed identification method on two classical systems in biology and physics, namely, a genetic repressilator network and a large scale network of interconnected Kuramoto oscillators.

We generalize the recently proposed Alternating Anderson-Jacobi (AAJ) method (Pratapa et al., J. Comput. Phys. (2016), 306, 43--54) to include preconditioning, and demonstrate its efficiency and scaling in the solution of large, sparse linear systems on parallel computers. The resulting preconditioned Alternating Anderson-Richardson (AAR) method reduces to the AAJ method for a particular choice of preconditioner. The AAR method employs Anderson extrapolation at periodic intervals within a preconditioned Richardson iteration to accelerate convergence. In this work, we develop a version of the method that is particularly well suited for scalable high-performance computing. In applications to Helmholtz and Poisson equations, we show that the strong and weak parallel scaling of AAR is superior to both Generalized Minimal Residual (GMRES) and Conjugate Gradient (CG) methods, using the same preconditioning, in large-scale parallel calculations employing up to 110,592 computational cores. Moreover, we find that the minimum time to solution for AAR is up to a factor of two smaller than both GMRES and CG. Overall, we find that the AAR method provides an efficient and scalable alternative to preconditioned Krylov solvers for the solution of large, sparse linear systems on high performance computing platforms.

Research conducted by the principal investigator during the period June 1, 1979 to May 31, 1980, resulted in the following research articles which have either appeared in print, or have been accepted in refereed mathematical journals, in this period: Inequalities for polynomials with a prescribed zero; On the Enestrom-Kakeya Theorem and Its Sharpness; Bounds for incomplete polynomials vanishing at both endpoints of an interval; Remarks on some conjectures of G. G. Lorentz; On incomplete polynomials; An extension of the Enestrom-Kakeya Theorem and its sharpness; On zeros of generalized Bessel polynomials; Incomplete factorizations of matrices and connections with H- matrices; On two conjectures on the zeros of generalized Bessel polynomials; Incomplete polynomials: an electrostatic approach; An introduction to the Convergence Theory of Pade' Approximants.

We discuss how inference can be performed when data are sampled from the non-ergodic phase of systems with multiple attractors. We take as model system the finite connectivity Hopfield model in the memory phase and suggest a cavity method approach to reconstruct the couplings when the data are separately sampled from few attractor states. We also show how the inference results can be converted into a learning protocol for neural networks in which patterns are presented through weak external fields. The protocol is simple and fully local, and is able to store patterns with a finite overlap with the input patterns without ever reaching a spin glass phase where all memories are lost.

Three properties of matrices: the spark, the mutual incoherence and the restricted isometry property have recently been introduced in the context of compressed sensing. We study these properties for matrices that are Kronecker products and show how these properties relate to those of the factors. For the mutual incoherence we also discuss results for sums of Kronecker products.