Downloads & Free Reading Options - Results

Finding inclusion-minimal "hitting sets" for a given collection of sets is a fundamental combinatorial problem with applications in domains as diverse as Boolean algebra, computational biology, and data mining. Much of the algorithmic literature focuses on the problem of *recognizing* the collection of minimal hitting sets; however, in many of the applications, it is more important to *generate* these hitting sets. We survey twenty algorithms from across a variety of domains, considering their history, classification, useful features, and computational performance on a variety of synthetic and real-world inputs. We also provide a suite of implementations of these algorithms with a ready-to-use, platform-agnostic interface based on Docker containers and the AlgoRun framework, so that interested computational scientists can easily perform similar tests with inputs from their own research areas on their own computers or through a convenient Web interface.

The goal of this research is to develop high-performance computational tools for designing control systems for a class of complex physical systems. (1) Developed a verification algorithm for verifying control laws using phase-space geometric modeling of dynamical systems. The algorithm evolves a hierarchically-refined bound of system nonlinear dynamics and can address practical concerns such as sensor, actuator, and modeling uncertainties in a systematic manner. We have applied the algorithm to the maglev control system prototype and compared the results against the physical measurements. (2) Constructed a physical experiment to study distributed acoustic sensing. The experiment comprises an enclosed chamber measured 1.7m x 0.846m x 0.201m, and an 8-channel A/D and D/A system with 6 microphones and a speaker. (3) Started to investigate control system design and optimization for distributed parameter physical systems (systems modeled by partial differential equations). (4) Will present the control verification paper at the IFAC International Symposium on Artificial Intelligence in Real-Time Control, Grand Canyon, AZ in October. Have presented an invited tutorial at AAAI National Conference in Madison, Wisconsin, July 1998. Jeff May and Feng Zhao, "Verification of control laws using phase-space geometric modeling of dynamical systems." IFAC AIRTC-98. Feng Zhao and Chris Bailey-Kellogg, "Intelligent Simulation." AAAI-98 Tutorial Forum.

Algorithms and Computation: 12th International Symposium, ISAAC 2001 Christchurch, New Zealand, December 19–21, 2001 Proceedings Author: Peter Eades, Tadao Takaoka Published by Springer Berlin Heidelberg ISBN: 978-3-540-42985-2 DOI: 10.1007/3-540-45678-3 Table of Contents: Chain Reconfiguration The Ins and Outs, Ups and Downs of Moving Polygons and Polygonal Linkages Application of M-Convex Submodular Flow Problem to Mathematical Economics A Polynomial Time Approximation Scheme for Minimizing Total Completion Time of Unbounded Batch Scheduling A Polynomial Time Approximation Scheme for the Multi-vehicle Scheduling Problem on a Path with Release and Handling Times Semi-normal Schedulings: Improvement on Goemans’ Algorithm Balanced Scheduling toward Loss-Free Packet Queuing and Delay Fairness Broadcasting with Universal Lists Revisited: Using Competitive Analysis On Adaptive Fault Diagnosis for Multiprocessor Systems On-Line Multicasting in All-Optical Networks Enumerating Floorplans with n Rooms On Min-Max Cycle Bases On the Minimum Local-Vertex-Connectivity Augmentation in Graphs Counting Graph Isomorphisms among Chordal Graphs with Restricted Clique Number Quantum Algorithms for Intersection and Proximity Problems BUSHWHACK: An Approximation Algorithm for Minimal Paths through Pseudo-Euclidean Spaces Approximation of Minimum Triangulation for Polyhedron with Bounded Degrees Tree-Approximations for the Weighted Cost-Distance Problem Necessary and Sufficient Numbers of Cards for Sharing Secret Keys on Hierarchical Groups Generic Algorithms and Key Agreement Protocols Based on Group Actions Baire Category and Nowhere Differentiability for Feasible Real Functions

Given an input $x$, and a search problem $F$, local computation algorithms (LCAs) implement access to specified locations of $y$ in a legal output $y \in F(x)$, using polylogarithmic time and space. Mansour et al., (2012), had previously shown how to convert certain online algorithms to LCAs. In this work, we expand on that line of work and develop new techniques for designing LCAs and bounding their space and time complexity. Our contributions are fourfold: (1) We significantly improve the running times and space requirements of LCAs for previous results, (2) we expand and better define the family of online algorithms which can be converted to LCAs using our techniques, (3) we show that our results apply to a larger family of graphs than that of previous results, and (4) our proofs are simpler and more concise than the previous proof methods. For example, we show how to construct LCAs that require $O(\log{n}\log\log{n})$ space and $O(\log^2{n})$ time (and expected time $O(\log\log{n})$) for problems such as maximal matching on a large family of graphs, as opposed to the henceforth best results that required $O(\log^3{n})$ space and $O(\log^4{n})$ time, and applied to a smaller family of graphs.

Research results include: (1) Developed empirical performance criteria for characterizing stabilities and robustness of the maglev control experimental system and completed a preliminary implementation of the performance characterization algorithms; (2) Developed and experimented with a phase-space search algorithm for synthesizing control actions; (3) Presented part of the above results in an article, 'Phas-Space Nonlinear Control Tool-box: The Maglev Experience' at HS'97: Fifth International Hybrid Systems Workshop, Notre Dame, IN, Sept. 11-13,1997

Research results include: (1) Developed empirical performance criteria for characterizing stabilities and robustness of the maglev control experimental system and completed a preliminary implementation of the performance characterization algorithms; (2) Developed and experimented with a phase-space search algorithm for synthesizing control actions; (3) Presented part of the above results in an article, 'Phas-Space Nonlinear Control Tool-box: The Maglev Experience' at HS'97: Fifth International Hybrid Systems Workshop, Notre Dame, IN, Sept. 11-13,1997

Research results include: (1) Developed empirical performance criteria for characterizing stabilities and robustness of the maglev control experimental system and completed a preliminary implementation of the performance characterization algorithms; (2) Developed and experimented with a phase-space search algorithm for synthesizing control actions; (3) Presented part of the above results in an article, 'Phas-Space Nonlinear Control Tool-box: The Maglev Experience' at HS'97: Fifth International Hybrid Systems Workshop, Notre Dame, IN, Sept. 11-13,1997

Research results include: (1) Developed empirical performance criteria for characterizing stabilities and robustness of the maglev control experimental system and completed a preliminary implementation of the performance characterization algorithms; (2) Developed and experimented with a phase-space search algorithm for synthesizing control actions; (3) Presented part of the above results in an article, 'Phas-Space Nonlinear Control Tool-box: The Maglev Experience' at HS'97: Fifth International Hybrid Systems Workshop, Notre Dame, IN, Sept. 11-13,1997

The driving force in the pursuit for quantum computation is the exciting possibility that quantum algorithms can be more efficient than their classical analogues. Research on the subject has unraveled several aspects of how that can happen. Clever quantum algorithms have been discovered in recent years, although not systematically, and the field remains under active investigation. Richard Feynman was one of the pioneers who foresaw the power of quantum computers. In this issue dedicated to him, I give an introduction to how particle and wave aspects contribute to the power of quantum computers. Shor's and Grover's algorithms are analysed as examples.

As a Guest Computational Investigator under the NASA administered component of the High Performance Computing and Communication Program, we implemented a massively parallel genetic algorithm on the MasPar SIMD computer. Experiments were conducted using Earth Science data in the domains of meteorology and oceanography. Results obtained in these domains are competitive with, and in most cases better than, similar problems solved using other methods. In the meteorological domain, we chose to identify clouds using AVHRR spectral data. Four cloud speciations were used although most researchers settle for three. Results were remarkedly consistent across all tests (91% accuracy). Refinements of this method may lead to more timely and complete information for Global Circulation Models (GCMS) that are prevalent in weather forecasting and global environment studies. In the oceanographic domain, we chose to identify ocean currents from a spectrometer having similar characteristics to AVHRR. Here the results were mixed (60% to 80% accuracy). Given that one is willing to run the experiment several times (say 10), then it is acceptable to claim the higher accuracy rating. This problem has never been successfully automated. Therefore, these results are encouraging even though less impressive than the cloud experiment. Successful conclusion of an automated ocean current detection system would impact coastal fishing, naval tactics, and the study of micro-climates. Finally we contributed to the basic knowledge of GA (genetic algorithm) behavior in parallel environments. We developed better knowledge of the use of subpopulations in the context of shared breeding pools and the migration of individuals. Rigorous experiments were conducted based on quantifiable performance criteria. While much of the work confirmed current wisdom, for the first time we were able to submit conclusive evidence. The software developed under this grant was placed in the public domain. An extensive user's manual was written and distributed nationwide to scientists whose work might benefit from its availability. Several papers, including two journal articles, were produced.

The materials in this teacher's guide are designed for about 18 weeks of study by secondary school students. For maximum benefit, the student needs contact with a computer, primarily for verifying and trouble-shooting the algorithms which he or she has constructed. The course is usually taught for grade 11 or 12 students. The commentary contains background material, suggestions for use, and answers for exercises for each chapter of the student text. Comments indicate the course requires more preparation time for the teacher than most high school mathematics courses; use of a student assistant is recommended. (RH)

Reference books for data structures and algorithm research.

Carrier-grade networks comprise several layers where different protocols coexist. Nowadays, most of these networks have different control planes to manage routing on different layers, leading to a suboptimal use of the network resources and additional operational costs. However, some routers are able to encapsulate, decapsulate and convert protocols and act as a liaison between these layers. A unified control plane would be useful to optimize the use of the network resources and automate the routing configurations. Software-Defined Networking (SDN) based architectures, such as OpenFlow, offer a chance to design such a control plane. One of the most important problems to deal with in this design is the path computation process. Classical path computation algorithms cannot resolve the problem as they do not take into account encapsulations and conversions of protocols. In this paper, we propose algorithms to solve this problem and study several cases: Path computation without bandwidth constraint, under bandwidth constraint and under other Quality of Service constraints. We study the complexity and the scalability of our algorithms and evaluate their performances on real topologies. The results show that they outperform the previous ones proposed in the literature.

We provide faster algorithms and improved sample complexities for approximating the top eigenvector of a matrix. Offline Setting: Given an $n \times d$ matrix $A$, we show how to compute an $\epsilon$ approximate top eigenvector in time $\tilde O ( [nnz(A) + \frac{d \cdot sr(A)}{gap^2}]\cdot \log 1/\epsilon )$ and $\tilde O([\frac{nnz(A)^{3/4} (d \cdot sr(A))^{1/4}}{\sqrt{gap}}]\cdot \log1/\epsilon )$. Here $sr(A)$ is the stable rank and $gap$ is the multiplicative eigenvalue gap. By separating the $gap$ dependence from $nnz(A)$ we improve on the classic power and Lanczos methods. We also improve prior work using fast subspace embeddings and stochastic optimization, giving significantly improved dependencies on $sr(A)$ and $\epsilon$. Our second running time improves this further when $nnz(A) \le \frac{d\cdot sr(A)}{gap^2}$. Online Setting: Given a distribution $D$ with covariance matrix $\Sigma$ and a vector $x_0$ which is an $O(gap)$ approximate top eigenvector for $\Sigma$, we show how to refine to an $\epsilon$ approximation using $\tilde O(\frac{v(D)}{gap^2} + \frac{v(D)}{gap \cdot \epsilon})$ samples from $D$. Here $v(D)$ is a natural variance measure. Combining our algorithm with previous work to initialize $x_0$, we obtain a number of improved sample complexity and runtime results. For general distributions, we achieve asymptotically optimal accuracy as a function of sample size as the number of samples grows large. Our results center around a robust analysis of the classic method of shift-and-invert preconditioning to reduce eigenvector computation to approximately solving a sequence of linear systems. We then apply fast SVRG based approximate system solvers to achieve our claims. We believe our results suggest the general effectiveness of shift-and-invert based approaches and imply that further computational gains may be reaped in practice.

This note is concerned with the explicit symbolic computation of expressions involving differential operators and their actions on functions. The derivation of specialized numerical algorithms, the explicit symbolic computation of integrals of motion, and the explicit computation of normal forms for nonlinear systems all require such computations. More precisely, if R = k(x(sub 1),...,x(sub N)), where k = R or C, F denotes a differential operator with coefficients from R, and g member of R, we describe data structures and algorithms for efficiently computing g. The basic idea is to impose a multiplicative structure on the vector space with basis the set of finite rooted trees and whose nodes are labeled with the coefficients of the differential operators. Cancellations of two trees with r + 1 nodes translates into cancellation of O(N(exp r)) expressions involving the coefficient functions and their derivatives.

This note is concerned with the explicit symbolic computation of expressions involving differential operators and their actions on functions. The derivation of specialized numerical algorithms, the explicit symbolic computation of integrals of motion, and the explicit computation of normal forms for nonlinear systems all require such computations. More precisely, if R = k(x(sub 1),...,x(sub N)), where k = R or C, F denotes a differential operator with coefficients from R, and g member of R, we describe data structures and algorithms for efficiently computing g. The basic idea is to impose a multiplicative structure on the vector space with basis the set of finite rooted trees and whose nodes are labeled with the coefficients of the differential operators. Cancellations of two trees with r + 1 nodes translates into cancellation of O(N(exp r)) expressions involving the coefficient functions and their derivatives.

This note is concerned with the explicit symbolic computation of expressions involving differential operators and their actions on functions. The derivation of specialized numerical algorithms, the explicit symbolic computation of integrals of motion, and the explicit computation of normal forms for nonlinear systems all require such computations. More precisely, if R = k(x(sub 1),...,x(sub N)), where k = R or C, F denotes a differential operator with coefficients from R, and g member of R, we describe data structures and algorithms for efficiently computing g. The basic idea is to impose a multiplicative structure on the vector space with basis the set of finite rooted trees and whose nodes are labeled with the coefficients of the differential operators. Cancellations of two trees with r + 1 nodes translates into cancellation of O(N(exp r)) expressions involving the coefficient functions and their derivatives.

Parallel computation of manipulator forward dynamics is investigated. Considering three classes of algorithms for the solution of the problem, that is, the O(n), the O(n exp 2), and the O(n exp 3) algorithms, parallelism in the problem is analyzed. It is shown that the problem belongs to the class of NC and that the time and processors bounds are of O(log2/2n) and O(n exp 4), respectively. However, the fastest stable parallel algorithms achieve the computation time of O(n) and can be derived by parallelization of the O(n exp 3) serial algorithms. Parallel computation of the O(n exp 3) algorithms requires the development of parallel algorithms for a set of fundamentally different problems, that is, the Newton-Euler formulation, the computation of the inertia matrix, decomposition of the symmetric, positive definite matrix, and the solution of triangular systems. Parallel algorithms for this set of problems are developed which can be efficiently implemented on a unique architecture, a triangular array of n(n+2)/2 processors with a simple nearest-neighbor interconnection. This architecture is particularly suitable for VLSI and WSI implementations. The developed parallel algorithm, compared to the best serial O(n) algorithm, achieves an asymptotic speedup of more than two orders-of-magnitude in the computation the forward dynamics.

Parallel computation of manipulator forward dynamics is investigated. Considering three classes of algorithms for the solution of the problem, that is, the O(n), the O(n exp 2), and the O(n exp 3) algorithms, parallelism in the problem is analyzed. It is shown that the problem belongs to the class of NC and that the time and processors bounds are of O(log2/2n) and O(n exp 4), respectively. However, the fastest stable parallel algorithms achieve the computation time of O(n) and can be derived by parallelization of the O(n exp 3) serial algorithms. Parallel computation of the O(n exp 3) algorithms requires the development of parallel algorithms for a set of fundamentally different problems, that is, the Newton-Euler formulation, the computation of the inertia matrix, decomposition of the symmetric, positive definite matrix, and the solution of triangular systems. Parallel algorithms for this set of problems are developed which can be efficiently implemented on a unique architecture, a triangular array of n(n+2)/2 processors with a simple nearest-neighbor interconnection. This architecture is particularly suitable for VLSI and WSI implementations. The developed parallel algorithm, compared to the best serial O(n) algorithm, achieves an asymptotic speedup of more than two orders-of-magnitude in the computation the forward dynamics.

Parallel computation of manipulator forward dynamics is investigated. Considering three classes of algorithms for the solution of the problem, that is, the O(n), the O(n exp 2), and the O(n exp 3) algorithms, parallelism in the problem is analyzed. It is shown that the problem belongs to the class of NC and that the time and processors bounds are of O(log2/2n) and O(n exp 4), respectively. However, the fastest stable parallel algorithms achieve the computation time of O(n) and can be derived by parallelization of the O(n exp 3) serial algorithms. Parallel computation of the O(n exp 3) algorithms requires the development of parallel algorithms for a set of fundamentally different problems, that is, the Newton-Euler formulation, the computation of the inertia matrix, decomposition of the symmetric, positive definite matrix, and the solution of triangular systems. Parallel algorithms for this set of problems are developed which can be efficiently implemented on a unique architecture, a triangular array of n(n+2)/2 processors with a simple nearest-neighbor interconnection. This architecture is particularly suitable for VLSI and WSI implementations. The developed parallel algorithm, compared to the best serial O(n) algorithm, achieves an asymptotic speedup of more than two orders-of-magnitude in the computation the forward dynamics.

Parallel computation of manipulator forward dynamics is investigated. Considering three classes of algorithms for the solution of the problem, that is, the O(n), the O(n exp 2), and the O(n exp 3) algorithms, parallelism in the problem is analyzed. It is shown that the problem belongs to the class of NC and that the time and processors bounds are of O(log2/2n) and O(n exp 4), respectively. However, the fastest stable parallel algorithms achieve the computation time of O(n) and can be derived by parallelization of the O(n exp 3) serial algorithms. Parallel computation of the O(n exp 3) algorithms requires the development of parallel algorithms for a set of fundamentally different problems, that is, the Newton-Euler formulation, the computation of the inertia matrix, decomposition of the symmetric, positive definite matrix, and the solution of triangular systems. Parallel algorithms for this set of problems are developed which can be efficiently implemented on a unique architecture, a triangular array of n(n+2)/2 processors with a simple nearest-neighbor interconnection. This architecture is particularly suitable for VLSI and WSI implementations. The developed parallel algorithm, compared to the best serial O(n) algorithm, achieves an asymptotic speedup of more than two orders-of-magnitude in the computation the forward dynamics.

Parallel computation of manipulator forward dynamics is investigated. Considering three classes of algorithms for the solution of the problem, that is, the O(n), the O(n exp 2), and the O(n exp 3) algorithms, parallelism in the problem is analyzed. It is shown that the problem belongs to the class of NC and that the time and processors bounds are of O(log2/2n) and O(n exp 4), respectively. However, the fastest stable parallel algorithms achieve the computation time of O(n) and can be derived by parallelization of the O(n exp 3) serial algorithms. Parallel computation of the O(n exp 3) algorithms requires the development of parallel algorithms for a set of fundamentally different problems, that is, the Newton-Euler formulation, the computation of the inertia matrix, decomposition of the symmetric, positive definite matrix, and the solution of triangular systems. Parallel algorithms for this set of problems are developed which can be efficiently implemented on a unique architecture, a triangular array of n(n+2)/2 processors with a simple nearest-neighbor interconnection. This architecture is particularly suitable for VLSI and WSI implementations. The developed parallel algorithm, compared to the best serial O(n) algorithm, achieves an asymptotic speedup of more than two orders-of-magnitude in the computation the forward dynamics.

This text contains material designed for about 18 weeks of study at grades 11 or 12. Use of a computer with the course is highly recommended. Developing an understanding of the relationship between mathematics, computers, and problem solving is the main objective of this book. The following chapters are included in the book: (1) Algorithms, Language, and Machines; (2) Input, Output, and Assignment; (3) Branching and Subscripted Variables; (4) Looping; (5) Functions and Procedures; (6) Approximations; (7) Some Mathematical Applications; and (8) Compilation and Some Other Non-Numeric Problems. Also included is a discussion on future computer applications. (RH)

We give two algorithms computing representative families of linear and uniform matroids and demonstrate how to use representative families for designing single-exponential parameterized and exact exponential time algorithms. The applications of our approach include - LONGEST DIRECTED CYCLE - MINIMUM EQUIVALENT GRAPH (MEG) - Algorithms on graphs of bounded treewidth -k-PATH, k-TREE, and more generally, k-SUBGRAPH ISOMORPHISM, where the k-vertex pattern graph is of constant treewidth.

Betweenness Centrality (BC) is steadily growing in popularity as a metrics of the influence of a vertex in a graph. The BC score of a vertex is proportional to the number of all-pairs-shortest-paths passing through it. However, complete and exact BC computation for a large-scale graph is an extraordinary challenge that requires high performance computing techniques to provide results in a reasonable amount of time. Our approach combines bi-dimensional (2-D) decomposition of the graph and multi-level parallelism together with a suitable data-thread mapping that overcomes most of the difficulties caused by the irregularity of the computation on GPUs. Furthermore, we propose novel heuristics which exploit the topology information of the graph in order to reduce time and space requirements of BC computation. Experimental results on synthetic and real-world graphs show that the proposed techniques allow the BC computation of graphs which are too large to fit in the memory of a single computational node along with a significant reduction of the computing time.

This is the student's textbook for Algorithms, Computation, and Mathematics (Fortran Supplement). This computer language supplement is split off from the main text to enable a school to choose the computer language desired, and also to make it easier to modify the course as languages change. The chapters in the text are designed to add language capability. Each can be read in conjunction with the main text, section by section. (RH)

Due to their small size and ever-increasing computing capability, mobile devices, such as smart phones, are ideal tools for computation in the battlefield. However, there are some mission-critical applications that cannot be completed in real-time on these devices. Some of these applications can meet their deadlines through computation offloading: sending computationally expensive operations to high-performance computers (HPCs). However, in battlefields it may not be possible to access traditional offloading targets, such as cloud computing services. In this project we examine computation offloading on mobile ad-hoc networks (MANETs) using vehicle-mounted HPCs as offload targets. We use the network simulator ns-3 to model computation offloading in the battlefield. We implemented a suite of extensible ns-3 models, which allow for a wide variety of experiments with MANETs. Using these models, we study a number of aspects of computation offloading including: HPC placement, client offloading strategies, effects of HPC mobility, and variations in network topology. In this report, we describe the overall architecture of the models and provide experimental results on the performance gained by computation offloading when a number of parameters are varied. Our initial results indicate that offloading to mobile HPCs increases the utility of mobile devices as computation platforms in the battlefield.

In the past several years, domain decomposition was a very popular topic, partly motivated by the potential of parallelization. While a large body of theory and algorithms were developed for model elliptic problems, they are only recently starting to be tested on realistic applications. The application of some of these methods to two model problems in computational fluid dynamics are investigated. Some examples are two dimensional convection-diffusion problems and the incompressible driven cavity flow problem. The construction and analysis of efficient preconditioners for the interface operator to be used in the iterative solution of the interface solution is described. For the convection-diffusion problems, the effect of the convection term and its discretization on the performance of some of the preconditioners is discussed. For the driven cavity problem, the effectiveness of a class of boundary probe preconditioners is discussed.

In this paper we present efficient algorithms for the computation of several invariant objects for Hamiltonian dynamics. More precisely, we consider KAM tori (i.e diffeomorphic copies of the torus such that the motion on them is conjugated to a rigid rotation) both Lagrangian tori (of maximal dimension) and whiskered tori (i.e. tori with hyperbolic directions which, together with the tangents to the torus and the symplectic conjugates span the whole tangent space). In the case of whiskered tori, we also present algorithms to compute the invariant splitting and the invariant manifolds associated to the splitting. We present them both for the case of discrete time and for differential equations. The algorithms are based on a Newton method to solve an appropriately chosen functional equation that expresses invariance. The algorithms are efficient: if we discretize the objects by $N$ elements, one step of the Newton method requires only O(N) storage and $O(N \ln(N))$ operations. Furthermore, if the object we consider is of dimension $\ell$, we only need to compute functions of $\ell$ variables, independently of what is the dimension of the phase space. The algorithms do not require that the system is presented in action-angle variables nor that it is close to integrable. The algorithms are backed up by rigorous \emph{a-posteriori} bounds which state that if the equations are solved with a small residual and some explicitly computable condition numbers are not too big, then, there is a true solution which is close to the computed one. The algorithms apply both to primary (i.e non-contractible) and secondary tori (i.e. contractible to a torus of lower dimension, such as islands). They have already been implemented. We will report on the technicalities of the implementation and the results of running them elsewhere.

xix,298p

We discuss new global optimization algorithms that are related to the stochastic methods of Rinnooy Kan and Timmer, and to our previous static, synchronous parallel version of this method. The new algorithms have two main new features. First, they adaptively concentrate the computation in the areas of the domain space that appear most likely to produce the global minimum. Secondly, on parallel computers, they use an asynchronous approach, combined with a central work scheduler, to avoid load balancing problems. We investigate several mechanisms for deciding when and how to make the adaptive adjustments. We also describe both algorithmic and implementation considerations involved in constructing the parallel asynchronous algorithm. Computational tests on sequential and parallel computers show that the adaptive and asynchronous features of our new method can substantially reduce the number of function evaluations, and the execution time, required by previous stochastic methods to solve global optimization problems.

This is the teacher's guide and commentary for the SMSG textbook Algorithms, Computation and Mathematics (Algol Supplement). This teacher's commentary provides background information for the teacher, suggestions for activities found in the student's Algol Supplement, and answers to exercises and activities. The course is designed for high school students in grades 11 and 12. Access to a computer is highly recommended. (RH)

We discuss the relation between the fluctuation-dissipation relation derived by Chatelain and Ricci-Tersenghi [C.Chatelain, J.Phys. A {\bf 36}, 10739 (2003); F. Ricci-Tersenghi, Phys.Rev.E 68, 065104(R) (2003)] and that by Lippiello-Corberi-Zannetti [E. Lippiello, F. Corberi and M. Zannetti Phys. Rev. E {\bf 72}, 056103 (2005)]. In order to do that, we re-derive the fluctuation-dissipation relation for systems of discrete variables evolving in discrete time via a stochastic non-equilibrium Markov process. The calculation is carried out in a general formalism comprising the Chatelain, Ricci-Tersenghi result and that by Lippiello-Corberi-Zannetti as special cases. The applicability, generality, and experimental feasibility of the two approaches is thoroughly discussed. Extending the analytical calculation to the variance of the response function we show the vantage of field-free numerical methods with respect to the standard method where the perturbation is applied. We also show that the signal to noise ratio is better (by a factor $\sqrt 2$) in the algorithm of Lippiello-Corberi-Zannetti with respect to that of Chatelain-Ricci Tersenghi.

This paper considers the problem of canonical-correlation analysis (CCA) (Hotelling, 1936) and, more broadly, the generalized eigenvector problem for a pair of symmetric matrices. These are two fundamental problems in data analysis and scientific computing with numerous applications in machine learning and statistics (Shi and Malik, 2000; Hardoon et al., 2004; Witten et al., 2009). We provide simple iterative algorithms, with improved runtimes, for solving these problems that are globally linearly convergent with moderate dependencies on the condition numbers and eigenvalue gaps of the matrices involved. We obtain our results by reducing CCA to the top-$k$ generalized eigenvector problem. We solve this problem through a general framework that simply requires black box access to an approximate linear system solver. Instantiating this framework with accelerated gradient descent we obtain a running time of $O(\frac{z k \sqrt{\kappa}}{\rho} \log(1/\epsilon) \log \left(k\kappa/\rho\right))$ where $z$ is the total number of nonzero entries, $\kappa$ is the condition number and $\rho$ is the relative eigenvalue gap of the appropriate matrices. Our algorithm is linear in the input size and the number of components $k$ up to a $\log(k)$ factor. This is essential for handling large-scale matrices that appear in practice. To the best of our knowledge this is the first such algorithm with global linear convergence. We hope that our results prompt further research and ultimately improve the practical running time for performing these important data analysis procedures on large data sets.

Let $K$ be a field equipped with a valuation. Tropical varieties over $K$ can be defined with a theory of Gr\"obner bases taking into account the valuation of $K$. Because of the use of the valuation, this theory is promising for stable computations over polynomial rings over a $p$-adic fields.We design a strategy to compute such tropical Gr\"obner bases by adapting the Matrix-F5 algorithm. Two variants of the Matrix-F5 algorithm, depending on how the Macaulay matrices are built, are available to tropical computation with respective modifications. The former is more numerically stable while the latter is faster.Our study is performed both over any exact field with valuation and some inexact fields like $\mathbb{Q}\_p$ or $\mathbb{F}\_q \llbracket t \rrbracket.$ In the latter case, we track the loss in precision, and show that the numerical stability can compare very favorably to the case of classical Gr\"obner bases when the valuation is non-trivial. Numerical examples are provided.

This is the teacher's guide and commentary for the SMSG textbook Algorithms, Computation, and Mathematics (Fortran Supplement). The teacher's commentary provides background information for the teacher, suggestions for activities found in the Fortran Supplement, and answers for exercises and activities. The course is designed for high school students in grades 11 and 12. Access to a computer is highly recommended. (RH)

This is the teacher's guide and commentary for the SMSG textbook Algorithms, Computation, and Mathematics (Fortran Supplement). The teacher's commentary provides background information for the teacher, suggestions for activities found in the Fortran Supplement, and answers for exercises and activities. The course is designed for high school students in grades 11 and 12. Access to a computer is highly recommended. (RH)

We give a deterministic 2^{O(n)} algorithm for computing an M-ellipsoid of a convex body, matching a known lower bound. This has several interesting consequences including improved deterministic algorithms for volume estimation of convex bodies and the shortest and closest lattice vector problems under general norms.

In general, work has continued on developing and evaluating algorithms designed to manage the Laser Atmospheric Wind Sounder (LAWS) lidar pulses and to compute the horizontal wind vectors from the line-of-sight (LOS) measurements. These efforts fall into three categories: Improvements to the shot management and multi-pair algorithms (SMA/MPA); observing system simulation experiments; and ground-based simulations of LAWS.

We consider optimal distributed computation of a given function of distributed data. The input (data) nodes and the sink node that receives the function form a connected network that is described by an undirected weighted network graph. The algorithm to compute the given function is described by a weighted directed acyclic graph and is called the computation graph. An embedding defines the computation communication sequence that obtains the function at the sink. Two kinds of optimal embeddings are sought, the embedding that---(1)~minimizes delay in obtaining function at sink, and (2)~minimizes cost of one instance of computation of function. This abstraction is motivated by three applications---in-network computation over sensor networks, operator placement in distributed databases, and module placement in distributed computing. We first show that obtaining minimum-delay and minimum-cost embeddings are both NP-complete problems and that cost minimization is actually MAX SNP-hard. Next, we consider specific forms of the computation graph for which polynomial time solutions are possible. When the computation graph is a tree, a polynomial time algorithm to obtain the minimum delay embedding is described. Next, for the case when the function is described by a layered graph we describe an algorithm that obtains the minimum cost embedding in polynomial time. This algorithm can also be used to obtain an approximation for delay minimization. We then consider bounded treewidth computation graphs and give an algorithm to obtain the minimum cost embedding in polynomial time.

We consider optimal distributed computation of a given function of distributed data. The input (data) nodes and the sink node that receives the function form a connected network that is described by an undirected weighted network graph. The algorithm to compute the given function is described by a weighted directed acyclic graph and is called the computation graph. An embedding defines the computation communication sequence that obtains the function at the sink. Two kinds of optimal embeddings are sought, the embedding that---(1)~minimizes delay in obtaining function at sink, and (2)~minimizes cost of one instance of computation of function. This abstraction is motivated by three applications---in-network computation over sensor networks, operator placement in distributed databases, and module placement in distributed computing. We first show that obtaining minimum-delay and minimum-cost embeddings are both NP-complete problems and that cost minimization is actually MAX SNP-hard. Next, we consider specific forms of the computation graph for which polynomial time solutions are possible. When the computation graph is a tree, a polynomial time algorithm to obtain the minimum delay embedding is described. Next, for the case when the function is described by a layered graph we describe an algorithm that obtains the minimum cost embedding in polynomial time. This algorithm can also be used to obtain an approximation for delay minimization. We then consider bounded treewidth computation graphs and give an algorithm to obtain the minimum cost embedding in polynomial time.

We present elementary mappings between classical lattice models and quantum circuits. These mappings provide a general framework to obtain efficiently simulable quantum gate sets from exactly solvable classical models. For example, we recover and generalize the simulability of Valiant's match-gates by invoking the solvability of the free-fermion eight-vertex model. Our mappings furthermore provide a systematic formalism to obtain simple quantum algorithms to approximate partition functions of lattice models in certain complex-parameter regimes. For example, we present an efficient quantum algorithm for the six-vertex model as well as a 2D Ising-type model. We finally show that simulating our quantum algorithms on a classical computer is as hard as simulating universal quantum computation (i.e. BQP-complete).

We present elementary mappings between classical lattice models and quantum circuits. These mappings provide a general framework to obtain efficiently simulable quantum gate sets from exactly solvable classical models. For example, we recover and generalize the simulability of Valiant's match-gates by invoking the solvability of the free-fermion eight-vertex model. Our mappings furthermore provide a systematic formalism to obtain simple quantum algorithms to approximate partition functions of lattice models in certain complex-parameter regimes. For example, we present an efficient quantum algorithm for the six-vertex model as well as a 2D Ising-type model. We finally show that simulating our quantum algorithms on a classical computer is as hard as simulating universal quantum computation (i.e. BQP-complete).

In this report, relevant parts of the scattering theory for magnetic spheres are presented. Mass extinction coefficients, and the lognormal size distribution are defined. The theory and algorithms for integrating scattering parameters over size distributions are developed. The integrations are carried out in terms of dimensionless scattering, and size distribution parameters, which are simply related to the usual mass scattering coefficients. Fortran codes, which implement the algorithmic design, are presented, and examples of code use are given. Code listings are included. (AN)

In this report, relevant parts of the scattering theory for magnetic spheres are presented. Mass extinction coefficients, and the lognormal size distribution are defined. The theory and algorithms for integrating scattering parameters over size distributions are developed. The integrations are carried out in terms of dimensionless scattering, and size distribution parameters, which are simply related to the usual mass scattering coefficients. Fortran codes, which implement the algorithmic design, are presented, and examples of code use are given. Code listings are included. (AN)

In this report, relevant parts of the scattering theory for magnetic spheres are presented. Mass extinction coefficients, and the lognormal size distribution are defined. The theory and algorithms for integrating scattering parameters over size distributions are developed. The integrations are carried out in terms of dimensionless scattering, and size distribution parameters, which are simply related to the usual mass scattering coefficients. Fortran codes, which implement the algorithmic design, are presented, and examples of code use are given. Code listings are included. (AN)

In this report, relevant parts of the scattering theory for magnetic spheres are presented. Mass extinction coefficients, and the lognormal size distribution are defined. The theory and algorithms for integrating scattering parameters over size distributions are developed. The integrations are carried out in terms of dimensionless scattering, and size distribution parameters, which are simply related to the usual mass scattering coefficients. Fortran codes, which implement the algorithmic design, are presented, and examples of code use are given. Code listings are included. (AN)

In this report, relevant parts of the scattering theory for magnetic spheres are presented. Mass extinction coefficients, and the lognormal size distribution are defined. The theory and algorithms for integrating scattering parameters over size distributions are developed. The integrations are carried out in terms of dimensionless scattering, and size distribution parameters, which are simply related to the usual mass scattering coefficients. Fortran codes, which implement the algorithmic design, are presented, and examples of code use are given. Code listings are included. (AN)

In this report, relevant parts of the scattering theory for magnetic spheres are presented. Mass extinction coefficients, and the lognormal size distribution are defined. The theory and algorithms for integrating scattering parameters over size distributions are developed. The integrations are carried out in terms of dimensionless scattering, and size distribution parameters, which are simply related to the usual mass scattering coefficients. Fortran codes, which implement the algorithmic design, are presented, and examples of code use are given. Code listings are included. (AN)

New methods for accelerating the determination of basis trees and dual evaluators for distribution problems are compared with standard solution procedures in a computational study of a wide range of distribution problems of varying sizes and densities. Computer programs utilizing the new methods are tested for computational efficiency in an experiment involving four solution techniques, four start algorithms, and four change of basis criteria, thus affording an empirical determination not only of the merits of various procedures in isolation but also of their effectiveness in combination. The study discloses that the most efficient solution procedure arises by coupling a primal transportation algorithm (embodying the accelerated updating and pricing methods) with a version of the Row Minimum start rule and a modified first negative evaluator rule. The resulting method was found to improve upon the efficiency of general purpose algorithms (taken from standard computer packages) by a factor of 50 or better, and also improved upon a streamlined version of the SHARE out-of-kilter code by a factor of 3. The method's median solution time for solving 175 x 175 distribution problems on a CDC 6600 computer was 11.4 seconds with a range of 9 to 13 seconds.