A Conservative Scheme for Parallel Interval Narrowing

Size: px
Start display at page:

Download "A Conservative Scheme for Parallel Interval Narrowing"

Transcription

1 Letters, 74(3-4): , Elsevier Science. A Conservative Scheme for Parallel Interval Narrowing Laurent Granvilliers a and Gaétan Hains b a IRIN, Université de Nantes, B.P , F Nantes Cedex 3, France b LIFO, Université d Orléans, B.P. 6759, F Orléans Cedex 2, France Abstract An efficient parallel interval narrowing algorithm for solving numerical problems is designed, implemented and tested. Differences with the corresponding sequential algorithm are clearly stated. The algorithm s performance is analyzed in the Bulk-Synchronous Parallel (BSP) cost model which suggests speed-ups on highbandwidth architectures. Experimental results on a massively parallel machine Cray T3E-1200 validate the model and show the parallel algorithm s efficiency as well as its limitations. Key words: parallel algorithms, numerical algorithms, interval narrowing, constraint propagation, BSP model. 1 Introduction Parallel processing of numerical problems via interval constraints has been proposed as a general framework for high-performance numerical computation in [5]. Its two potential advantages over classical methods are 1. the guarantee of numerically correct answers through interval arithmetic and IEEE standard arithmetic and 2. the use of only two algorithms, search and constraint propagation by interval contractions. This simple logical structure offers much scope for optimizing and/or parallelizing the two core algorithms. The problem of parallel search is in itself interesting and difficult but purely combinatorial. Nevertheless, whatever the complexity of the search problems faced by numerical solvers they must optimize constraint propagation. Efficient constraint propagation is therefore critical for the success of parallel interval constraint solving. Although general algorithms are known for it [10], it Preprint submitted to Elsevier Preprint 26 May 2000

2 has been observed that the classical notion of parallel speed-up is not a correct measure of success for such algorithms. It is a non-trivial problem to schedule interval contractions in parallel while retaining a rate of convergence comparable to the best sequential algorithms [4]. This is due to a parallel decoupling phenomenon: convergence may be faster when two interval contractions are applied in sequence than in parallel. In this paper we demonstrate for the first time that important speed-ups are possible for constraint propagation by parallelizing the purely asynchronous, selection part of the best known sequential heuristics. Analysis in the BSP cost model predicted speed-up for high-bandwidth architectures and our tests have confirmed this on T3E and Sun bi-processor systems. Our conservative algorithm is thus a proof of feasibility for parallel interval constraint solving and its limitations define precise requirements for future algorithms. The rest of the article is structured as follows. The next section recalls the definition of a constraint satisfaction problem on numerical intervals and its sequential solution by interval narrowing. The parallel algorithm is then described with its theoretical performance analysis. Finally, experimental results are shown to confirm the possibility of speed-ups and highlight the current algorithm s limitations. 2 Interval narrowing A floating-point interval, called thereafter interval, is a connected set of reals whose lowest upper bound and greatest lower bound are floating-point numbers. The notation [a.. b] is used as shorthand for the set {x R a x b}. The symbol I denotes the set of intervals. In the following, a constraint c is an atomic formula built from a real-based structure and a set of real-valued variables. It defines a relation ρ c on reals. A constraint satisfaction problem, denoted thereafter CSP, is a tuple X, C, I made of a set of real-valued variables X = {x 1,...,x n }, a set of constraints C built from X and a vector of intervals I = (I 1,..., I n ) where I k is the domain of x k for every k {1,..., n}. Interval narrowing algorithms narrow the variables domains while preserving the CSP s solution set, i.e. the set of tuples of I 1... I n satisfying all the constraints. As described in [1], they implement strategies for applying narrowing operators (elementary solvers) associated with constraints. A narrowing operator for a constraint c is a contracting, monotonic, idempotent function nar : I n I n such that ρ c I 1 I n nar (I) for every I I n (correctness property). It is said to depend on the variables contained in c. 2

3 Table 1 Interval narrowing algorithm. IN (in Nars : set of narrowing operators; inout I : vector of intervals) begin queue Nars repeat Choose nar in queue % selection J nar (I) % contraction/narrowing if J I then % propagation queue queue {nar Nars i {1,...,n} : J i I i nar depends on x i } % nar has to be reinvoked since the modification of a domain % it depends on may invalidate the fixed-point property % of the narrowing operators in Nars \ queue I J endif queue queue \ {nar} until queue = or I 1 I n = end % nar is removed from queue since it is idempotent % by definition The work described here uses algorithms that compute a fixed-point of the input narrowing operators 1, taking advantage of the confluence property: given a CSP their output is necessarily the greatest common fixed-point of the narrowing operators included in the Cartesian product of initial domains whatever the chosen strategy (order of application of narrowing operators). The quality of such algorithms is measured by their speed of convergence on various benchmark CSP. The sequential algorithm we parallelize is called IN (for interval narrowing) and is presented in Table 1. It is an immediate adaptation of AlgorithmAC3 [9] and of the filtering algorithms used in interval constraint-based systems like CLP(BNR) or Prolog IV. It maintains a queue that satisfies the following invariant: the domains are necessarily a fixed-point of every narrowing operator not in that queue. If the queue becomes empty, a global fixed-point is reached and if the product of domains ever becomes empty, the CSP s inconsistency has been established. 3 Parallel interval narrowing Table 2 presents the parallel interval narrowing (PIN) algorithm. It realizes a variant of algorithm IN using the following selection-contraction heuristic. 1 The implementation of narrowing operators is beyond the scope of this paper, see [1]. 3

4 First, all operators are applied independently. While doing so, every operator s rate of contraction is computed. A reduced queue is then built with the best operators for which a fixed-point is not yet reached. An operator is part of the reduced queue if its rate of contraction is maximal for some interval. Finally, Algorithm IN is applied to reach a fixed-point for the reduced queue and the whole procedure is repeated. The idea of selection is inspired by the seminal work of Lhomme et al. [8] on the acceleration of sequential interval narrowing algorithms. It produces very fast sequential convergence [2] and its selection phase is trivial to parallelize. The algorithm presented in [8] performs its selection based on cycles detected in the inter-operator dependencies for IN s strategy. It selects, for a given variable and a given cycle, the most contracting operator in that cycle. By comparison, PIN s selection is more optimal because it acts on the set of all operators. PIN also eliminates the sequential (reference) strategy for cycle detection. Although PIN uses IN as a procedure, we will not evaluate its parallel speedup by comparison with IN but with 1-processor PIN. The comparison between 1-processor PIN and IN has been studied in [3] where the former is found to be often much faster and sometimes marginally slower, for an equivalent precision. This experimental result is reinforced by the following general argument. The overall number of operator contractions for IN is O(n 2 md) where n is the number of variables, m is the number of constraints and d is the number of floats in the largest domain. Indeed, removal of one value from a domain can add (nm 1) operators to the queue. In the worst case, this is repeated for every of the nd values in the CSP. A similar argument gives an O(n 2 md) for PIN. These observations have led us to design PIN as alternating between a purely parallel selection (search) phase and a purely sequential propagation phase. The parallel selection phase is well-balanced if the narrowing operators computations times are quite comparable. That is the case for the narrowing operators implementing hull consistency [5], but this property is not guaranteed in general. For instance, box consistency is implemented by a binary search whose length strongly depends on the domains [3]. Finally, there are problems where communication costs dominate and PIN slows down with the number of processors. For this reason, we have used the BSP cost model to estimate PIN s parallelism/communication tradeoff before testing the algorithm. The BSP execution model [12] represents a parallel computation on p processors as an alternating sequence of computations supersteps (p asynchronous computations) and communications supersteps (data exchanges between processors) with global synchronization. The BSP cost model estimates execution times as follows. A computation superstep takes as long as its longest sequential process, a global synchronization takes a fixed, system-dependent 4

5 Table 2 Parallel interval narrowing algorithm. PIN (in Nars : set of narrowing operators; in p : number of processors; inout I : vector of intervals) % the processors are numbered from 0 to p 1 begin IDvar {1,...,n} repeat % 1. Creation of the global queue on all processors % identifiers of modified domains queue {nar Nars i IDvar and nar depends on x i } L size (queue)/p % block size for local queues % 2. Asynchronous parallel selection of narrowing operators in parallel for pid = 0 to p 1 do localqueue queue[(pid L).. min((pid + 1) L 1,size (queue))] for all nar localqueue do % J: vector of vectors of intervals: Nars [1.. n] Jnar nar (I) % contraction % W: vector of vectors of reals: Nars [1.. n] % Estimation of the quality of nar: % For Jnar,i = [c,d] [a,b] = I i define % d(jnar,i, I i ) = (c a) + (b d) with roundings towards + Wnar distance (I,Jnar) = n i=1 d(j nar,i, I i ) endfor endparfor % 3. Parallel computation of contracted domains and best operator for each. % Multiple fold operations: final result (reducedqueue,k) on processor 0. % K: vector of intervals: [1.. n] K i nar queue J nar,i reducedqueue n {nar nar(i) i I i Wnar,i = max m Nars (W m,i )} i=1 % 4. Sequential computation of a fixed-point of the selected operators where pid = 0 do if K I then IN (reducedqueue, K) IDvar {i {1,...,n} K i I i } % new modified domains I K else IDvar % fixed-point reached endif done until IDvar = or I 1 I n = end 5

6 time L and a communication superstep is completed in time proportional to a factor h: the maximal number of bytes sent or received by a processor. A system-dependent constant g is multiplied by h to obtain the estimated communication time. We now outline the analysis of PIN s execution time. The enumeration below refer to numbered parts of the main repeat loop in Table 2. The goal of this analysis is to determine whether PIN executed on p 2 processors can provide any speed-up w.r.t a sequential execution of PIN. For this purpose, it is possible to ignore the sequential propagation time of step 4 and consider only steps 1,2 and Creation of a copy of the global queue by every processor in time O(Ng+ L) where N is the total number of operators. 2. Selection is a pure computation superstep which completes in time proportional to t o N/p where t o is the number of Flops necessary for an operator s application. Its value depends on the CSP and was taken into account by our numerical estimates. 3. Step 3 is made entirely of standard parallel fold operations and complete in time σ/p+(log k p)(l+(k 1)g) where σ is proportional to the number of operators per variable and k 2, the arity of the reduction algorithm is to be selected as a function of (g, p, L). Numerical evaluation of the above estimates for published (g, p, L) parameters 2 has led us to conclude that speed-up forpin required the best possible g values. We have used a Sun bi-processor and a T3E architecture whose values for this parameter (measured with BSP-Lib s bsp probe) are indeed excellent: between 2.4 and Experimental results Algorithm PIN was implemented in C and BSPlib v1.4 [6] and ported on two architectures supporting IEEE binary floating-point arithmetic [7]: a massivelyparallel system Cray T3E-1200 with 256 nodes (450MHz and 128 MB per n- ode) and a Sun UltraSparc workstation with 2 nodes (166MHz per node with a 128MB shared memory). The results for the Sun system are consistent with those for the T3E system and we will only mention the latter. We report the results obtained for the dense system of nonlinear equations Moré-Cosnard [11] and the sparse system Broyden-Banded [2] well-known in 2 J.M.D.Hill, BSP cost parameters for the Oxford BSP toolset, Sept

7 Table 3 Experimental results on the Cray T3E (figures in seconds). n IN p= Seq Moré-Cosnard Moré-Cosnard Moré-Cosnard Broyden-Banded Broyden-Banded Broyden-Banded the Interval Analysis community. Both are square and scalable (of parameterized size), and they permit illustrating the extreme behaviours of Algorithms IN and PIN. Moré-Cosnard s equations are the following: x k + 1[(1 t 2 k) k j=1 t j (x j + t j + 1) 3 + t mj=k+1 k (1 t j )(x j + t j + 1) 2 ] = 0 1 k m t j = jh h = 1/(m + 1) x k [ 4..5] Broyden-Banded benchmark is expressed as follows: x k (2 + 5x 2 k ) + 1 j J k x j (1 + x j ) = 0 1 k m J k = {j j k & max(1, k 5) j min(m, k + 1)} x k [ ] Table 3 illustrate the total solving times on the Cray T3E of both Algorithms IN (column labelled by IN) and PIN. Figures in columns labelled by p = 1, 2,..., 64 are the solving times of PIN on the p-processors machine. The last column corresponds to the solving time thresholds of PIN, i.e. the computation times of the sequential process IN, that is independent from the number of processors. Both n-dimensional Moré-Cosnard and Broyden-Banded problems are tested; let remark that the number of narrowing operators for Moré- Cosnard is n 2, and about 7n for Broyden-Banded. The conclusions are the following: The selection phase is efficiently parallelized since its acceleration is almost linear according to the number of processors. Acceleration of solving depends on the amount of time spent during selection. Solving time is bounded by the computation time of Algorithm IN in PIN (column Seq). For instance, this time is about 7s and 3s respectively 7

8 for both Moré-Cosnard and Broyden-Banded problems with 40 variables. When the difference between selection and sequential solving is huge then total acceleration is small and even negative in the case of Broyden-banded: the communication times are greater than the gain due to parallel selection. The density of a system (the ratio between the number of constraints and the number of constraint-variable pairs) determines the number of narrowing operators generated and plays an important role in the distribution of the selection phase and in the global parallel speed-up. Algorithm PIN executed on one processor is more efficient than the pure sequential algorithm IN which is due to the selection phase of PIN. For example, the selection for Moré-Cosnard dismisses m 2 m narrowing operators from the queue, and about 7m m for Broyden-Banded since most variables appear in 7 constraints. In other words, the fixed-point is reached for both problems using only one narrowing operator per variable. The above conclusions are also valid for a diverse set of benchmarks used in [2] to demonstrate the capabilities of the selection-contraction strategy. Among them, 80 out of 100 exhibit important speed-ups. Moreover PIN is particularly effective for discovering fixed-points since in such a case only the parallel selection is executed, which process is well-balanced for the implemented narrowing operators computing box consistency [3], and fold operations have no datas to reduce. This case often happens in practice where constraint propagation alternates with a search process splitting the domains. 5 Conclusion We have studied theoretically and experimentally the possibility for speedups on systems of numerical interval constraints. The speed-ups observed for Algorithm PIN are relative to the best known sequential heuristics and are thus the lowest possible solution times for the given constraint systems. PIN s limitations follow from its centralized communication structure: the need for a high network bandwidth and an absolute lower-bound on solving time due to the sequential propagation part of the algorithm. Better parallel algorithms for CSP should make optimal use of distributed constraint propagation to improve on PIN s rate of convergence. Acknowledgments This work was supported by the Franco-Russian Liapunov project on real constraint solving and applications headed by Frédéric Benhamou and Tat yana Mikhajlovna Yakhno, and the LOCO project between INRIA Rocquencourt and University of Orléans. The authors thank the Central Institute for Applied Mathematics (ZAM) of Jülich, Germany for 8

9 making possible the experiments on the Cray T3E. G. Hains thanks Maarten van Emden for sharing many ideas related to parallel constraint solving. References [1] F. Benhamou. Heterogeneous Constraint Solving. In Proc. of Int. Conf. on Algebraic and Logic Programming, volume 1139 of LNCS, Aachen, Germany, [2] L. Granvilliers. Consistances locales et transformations symboliques de contraintes d intervalles. PhD thesis, Université d Orléans, France, [3] L. Granvilliers, F. Goualard, and F. Benhamou. Box Consistency through Weak Box Consistency. In Proc. of Int. Conf. on Tools with Artificial Intelligence, Chicago IL, USA, IEEE Computer Society. [4] L. Granvilliers, G. Hains, Q. Miller, and N. Romero. A System for the Highlevel Parallelization and Cooperation of Constraint Solvers. In Proc. of Int. Conf. on Parallel and Distributed Computing and Systems, Las Vegas, USA, [5] G. Hains and M. H. van Emden. Towards high-quality, high-speed numerical computation. In Proc. of Pacific Rim Conf. on Communications, Computers and Signal Processing, University of Victoria, B.C., Canada, IEEE. [6] J. M. D. Hill, B. McColl, D. C. Stefanescu, M. W. Goudreau, K. Lang, S. B. Rao, T. Suel, T. Tsantilas, and R. Bisseling. BSPlib: The BSP Programming Library. Technical report, Oxford University Computing Laboratory, [7] IEEE. IEEE Standard for Binary Floating-Point Arithmetic. Technical Report IEEE Std , IEEE, Reaffirmed [8] O. Lhomme, A. Gotlieb, and M. Rueher. Dynamic Optimization of Interval Narrowing Algorithms. Journal of Logic Programming, 37(1 2): , [9] A. Mackworth. Consistency in Networks of Relations. Artificial Intelligence, 8(1):99 118, [10] E. Monfroy and J-H. Réty. Chaotic Iteration for Distributed Constraint Propagation. In Proc. of ACM Symp. on Applied Computing, San Antonio, USA, [11] J. J. Moré and M. Y. Cosnard. Numerical Solution of Nonlinear Equations. ACM Transactions on Mathematical Software, 5:64 85, [12] L. Valiant. A Bridging Model for Parallel Computation. Communications of the ACM, pages ,

Consistency Techniques over Real Numbers

Consistency Techniques over Real Numbers Consistency Techniques over Real Numbers Laurent Granvilliers granvilliers@irin.univ-nantes.fr University of Nantes Laurent Granvilliers, Dagstuhl Seminar 2003 p. 1/20 Reduction and Propagation Compute

More information

MURDOCH RESEARCH REPOSITORY

MURDOCH RESEARCH REPOSITORY MURDOCH RESEARCH REPOSITORY http://dx.doi.org/10.1109/tencon.2000.893677 Xie, H. and Fung, C.C. (2000) Enhancing the performance of a BSP model-based parallel volume renderer with a profile visualiser.

More information

Global Optimization based on Contractor Programming: an Overview of the IBEX library

Global Optimization based on Contractor Programming: an Overview of the IBEX library Global Optimization based on Contractor Programming: an Overview of the IBEX library Jordan Ninin ENSTA-Bretagne, LabSTIC, IHSEV team, 2 rue Francois Verny, 29806 Brest, France, jordan.ninin@ensta-bretagne.fr

More information

Revising hull and box consistency

Revising hull and box consistency Appears in: Procs. of the 16th Intl. Conf. on Logic Programming, p. 230 244, The MIT Press, 1999 Revising hull and box consistency Frédéric Benhamou, Frédéric Goualard, Laurent Granvilliers IRIN, Université

More information

Unifying and extending hybrid tractable classes of CSPs

Unifying and extending hybrid tractable classes of CSPs Journal of Experimental & Theoretical Artificial Intelligence Vol. 00, No. 00, Month-Month 200x, 1 16 Unifying and extending hybrid tractable classes of CSPs Wady Naanaa Faculty of sciences, University

More information

Revisiting the Upper Bounding Process in a Safe Branch and Bound Algorithm

Revisiting the Upper Bounding Process in a Safe Branch and Bound Algorithm Revisiting the Upper Bounding Process in a Safe Branch and Bound Algorithm Alexandre Goldsztejn 1, Yahia Lebbah 2,3, Claude Michel 3, and Michel Rueher 3 1 CNRS / Université de Nantes 2, rue de la Houssinière,

More information

A CSP Search Algorithm with Reduced Branching Factor

A CSP Search Algorithm with Reduced Branching Factor A CSP Search Algorithm with Reduced Branching Factor Igor Razgon and Amnon Meisels Department of Computer Science, Ben-Gurion University of the Negev, Beer-Sheva, 84-105, Israel {irazgon,am}@cs.bgu.ac.il

More information

MANUFACTURING SIMULATION USING BSP TIME WARP WITH VARIABLE NUMBERS OF PROCESSORS

MANUFACTURING SIMULATION USING BSP TIME WARP WITH VARIABLE NUMBERS OF PROCESSORS MANUFACTURING SIMULATION USING BSP TIME WARP WITH VARIABLE NUMBERS OF PROCESSORS Malcolm Yoke Hean Low Programming Research Group, Computing Laboratory, University of Oxford Wolfson Building, Parks Road,

More information

arxiv:cs/ v1 [cs.ds] 11 May 2005

arxiv:cs/ v1 [cs.ds] 11 May 2005 The Generic Multiple-Precision Floating-Point Addition With Exact Rounding (as in the MPFR Library) arxiv:cs/0505027v1 [cs.ds] 11 May 2005 Vincent Lefèvre INRIA Lorraine, 615 rue du Jardin Botanique, 54602

More information

Validating Plans with Durative Actions via Integrating Boolean and Numerical Constraints

Validating Plans with Durative Actions via Integrating Boolean and Numerical Constraints Validating Plans with Durative Actions via Integrating Boolean and Numerical Constraints Roman Barták Charles University in Prague, Faculty of Mathematics and Physics Institute for Theoretical Computer

More information

A New Algorithm for Singleton Arc Consistency

A New Algorithm for Singleton Arc Consistency A New Algorithm for Singleton Arc Consistency Roman Barták, Radek Erben Charles University, Institute for Theoretical Computer Science Malostranské nám. 2/25, 118 Praha 1, Czech Republic bartak@kti.mff.cuni.cz,

More information

A Fast Arc Consistency Algorithm for n-ary Constraints

A Fast Arc Consistency Algorithm for n-ary Constraints A Fast Arc Consistency Algorithm for n-ary Constraints Olivier Lhomme 1 and Jean-Charles Régin 2 1 ILOG, 1681, route des Dolines, 06560 Valbonne, FRANCE 2 Computing and Information Science, Cornell University,

More information

General properties of staircase and convex dual feasible functions

General properties of staircase and convex dual feasible functions General properties of staircase and convex dual feasible functions JÜRGEN RIETZ, CLÁUDIO ALVES, J. M. VALÉRIO de CARVALHO Centro de Investigação Algoritmi da Universidade do Minho, Escola de Engenharia

More information

Conflict-based Statistics

Conflict-based Statistics Conflict-based Statistics Tomáš Müller 1, Roman Barták 1 and Hana Rudová 2 1 Faculty of Mathematics and Physics, Charles University Malostranské nám. 2/25, Prague, Czech Republic {muller bartak}@ktiml.mff.cuni.cz

More information

Parallel Greedy Graph Matching. Partitioning Approach

Parallel Greedy Graph Matching. Partitioning Approach using an Edge Partitioning Approach Md. Mostofa Ali Patwary* Rob H. Bisseling** Fredrik Manne* *Department of Informatics, University of Bergen, Norway **Department of Mathematics, Utrecht University,

More information

The Global Standard for Mobility (GSM) (see, e.g., [6], [4], [5]) yields a

The Global Standard for Mobility (GSM) (see, e.g., [6], [4], [5]) yields a Preprint 0 (2000)?{? 1 Approximation of a direction of N d in bounded coordinates Jean-Christophe Novelli a Gilles Schaeer b Florent Hivert a a Universite Paris 7 { LIAFA 2, place Jussieu - 75251 Paris

More information

Global Optimization based on Contractor Programming: an Overview of the IBEX library

Global Optimization based on Contractor Programming: an Overview of the IBEX library Global Optimization based on Contractor Programming: an Overview of the IBEX library Jordan Ninin To cite this version: Jordan Ninin. Global Optimization based on Contractor Programming: an Overview of

More information

Parallel Evaluation of Hopfield Neural Networks

Parallel Evaluation of Hopfield Neural Networks Parallel Evaluation of Hopfield Neural Networks Antoine Eiche, Daniel Chillet, Sebastien Pillement and Olivier Sentieys University of Rennes I / IRISA / INRIA 6 rue de Kerampont, BP 818 2232 LANNION,FRANCE

More information

Authors Abugchem, F. (Fathi); Short, M. (Michael); Xu, D. (Donglai)

Authors Abugchem, F. (Fathi); Short, M. (Michael); Xu, D. (Donglai) TeesRep - Teesside's Research Repository A Note on the Suboptimality of Nonpreemptive Real-time Scheduling Item type Article Authors Abugchem, F. (Fathi); Short, M. (Michael); Xu, D. (Donglai) Citation

More information

Bound Consistency for Binary Length-Lex Set Constraints

Bound Consistency for Binary Length-Lex Set Constraints Bound Consistency for Binary Length-Lex Set Constraints Pascal Van Hentenryck and Justin Yip Brown University, Box 1910 Carmen Gervet Boston University, Providence, RI 02912 808 Commonwealth Av. Boston,

More information

Efficient Algorithms for Functional Constraints

Efficient Algorithms for Functional Constraints Efficient Algorithms for Functional Constraints Yuanlin Zhang 1, Roland HC Yap 2, Chendong Li 1, and Satyanarayana Marisetti 1 1 Texas Tech University, USA 2 National University of Singapore, Singapore

More information

Feature clustering and mutual information for the selection of variables in spectral data

Feature clustering and mutual information for the selection of variables in spectral data Feature clustering and mutual information for the selection of variables in spectral data C. Krier 1, D. François 2, F.Rossi 3, M. Verleysen 1 1,2 Université catholique de Louvain, Machine Learning Group

More information

A more efficient algorithm for perfect sorting by reversals

A more efficient algorithm for perfect sorting by reversals A more efficient algorithm for perfect sorting by reversals Sèverine Bérard 1,2, Cedric Chauve 3,4, and Christophe Paul 5 1 Département de Mathématiques et d Informatique Appliquée, INRA, Toulouse, France.

More information

An algorithm for Performance Analysis of Single-Source Acyclic graphs

An algorithm for Performance Analysis of Single-Source Acyclic graphs An algorithm for Performance Analysis of Single-Source Acyclic graphs Gabriele Mencagli September 26, 2011 In this document we face with the problem of exploiting the performance analysis of acyclic graphs

More information

Hoard: A Fast, Scalable, and Memory-Efficient Allocator for Shared-Memory Multiprocessors

Hoard: A Fast, Scalable, and Memory-Efficient Allocator for Shared-Memory Multiprocessors Hoard: A Fast, Scalable, and Memory-Efficient Allocator for Shared-Memory Multiprocessors Emery D. Berger Robert D. Blumofe femery,rdbg@cs.utexas.edu Department of Computer Sciences The University of Texas

More information

A Randomized Algorithm for Minimizing User Disturbance Due to Changes in Cellular Technology

A Randomized Algorithm for Minimizing User Disturbance Due to Changes in Cellular Technology A Randomized Algorithm for Minimizing User Disturbance Due to Changes in Cellular Technology Carlos A. S. OLIVEIRA CAO Lab, Dept. of ISE, University of Florida Gainesville, FL 32611, USA David PAOLINI

More information

Graph Coloring on a Coarse Grained Multiprocessor

Graph Coloring on a Coarse Grained Multiprocessor Graph Coloring on a Coarse Grained Multiprocessor (extended abstract) Assefaw Hadish Gebremedhin 1, Isabelle Guérin Lassous 2, Jens Gustedt 3, and Jan Arne Telle 1 1 Univ. of Bergen, Norway. assefaw telle

More information

arxiv: v1 [cs.dm] 6 May 2009

arxiv: v1 [cs.dm] 6 May 2009 Solving the 0 1 Multidimensional Knapsack Problem with Resolution Search Sylvain Boussier a, Michel Vasquez a, Yannick Vimont a, Saïd Hanafi b and Philippe Michelon c arxiv:0905.0848v1 [cs.dm] 6 May 2009

More information

Some Applications of Graph Bandwidth to Constraint Satisfaction Problems

Some Applications of Graph Bandwidth to Constraint Satisfaction Problems Some Applications of Graph Bandwidth to Constraint Satisfaction Problems Ramin Zabih Computer Science Department Stanford University Stanford, California 94305 Abstract Bandwidth is a fundamental concept

More information

Theorem 2.9: nearest addition algorithm

Theorem 2.9: nearest addition algorithm There are severe limits on our ability to compute near-optimal tours It is NP-complete to decide whether a given undirected =(,)has a Hamiltonian cycle An approximation algorithm for the TSP can be used

More information

Throughout this course, we use the terms vertex and node interchangeably.

Throughout this course, we use the terms vertex and node interchangeably. Chapter Vertex Coloring. Introduction Vertex coloring is an infamous graph theory problem. It is also a useful toy example to see the style of this course already in the first lecture. Vertex coloring

More information

On the Max Coloring Problem

On the Max Coloring Problem On the Max Coloring Problem Leah Epstein Asaf Levin May 22, 2010 Abstract We consider max coloring on hereditary graph classes. The problem is defined as follows. Given a graph G = (V, E) and positive

More information

Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization

Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization Richa Agnihotri #1, Dr. Shikha Agrawal #1, Dr. Rajeev Pandey #1 # Department of Computer Science Engineering, UIT,

More information

Tabu Search for Constraint Solving and Its Applications. Jin-Kao Hao LERIA University of Angers 2 Boulevard Lavoisier Angers Cedex 01 - France

Tabu Search for Constraint Solving and Its Applications. Jin-Kao Hao LERIA University of Angers 2 Boulevard Lavoisier Angers Cedex 01 - France Tabu Search for Constraint Solving and Its Applications Jin-Kao Hao LERIA University of Angers 2 Boulevard Lavoisier 49045 Angers Cedex 01 - France 1. Introduction The Constraint Satisfaction Problem (CSP)

More information

On the Space-Time Trade-off in Solving Constraint Satisfaction Problems*

On the Space-Time Trade-off in Solving Constraint Satisfaction Problems* Appeared in Proc of the 14th Int l Joint Conf on Artificial Intelligence, 558-56, 1995 On the Space-Time Trade-off in Solving Constraint Satisfaction Problems* Roberto J Bayardo Jr and Daniel P Miranker

More information

Graph Coloring via Constraint Programming-based Column Generation

Graph Coloring via Constraint Programming-based Column Generation Graph Coloring via Constraint Programming-based Column Generation Stefano Gualandi Federico Malucelli Dipartimento di Elettronica e Informatica, Politecnico di Milano Viale Ponzio 24/A, 20133, Milan, Italy

More information

On partial order semantics for SAT/SMT-based symbolic encodings of weak memory concurrency

On partial order semantics for SAT/SMT-based symbolic encodings of weak memory concurrency On partial order semantics for SAT/SMT-based symbolic encodings of weak memory concurrency Alex Horn and Daniel Kroening University of Oxford April 30, 2015 Outline What s Our Problem? Motivation and Example

More information

Graph Partitioning for Scalable Distributed Graph Computations

Graph Partitioning for Scalable Distributed Graph Computations Graph Partitioning for Scalable Distributed Graph Computations Aydın Buluç ABuluc@lbl.gov Kamesh Madduri madduri@cse.psu.edu 10 th DIMACS Implementation Challenge, Graph Partitioning and Graph Clustering

More information

Binary Search. Roland Backhouse February 5th, 2001

Binary Search. Roland Backhouse February 5th, 2001 1 Binary Search Roland Backhouse February 5th, 2001 Outline 2 An implementation in Java of the card-searching algorithm is presented. Issues concerning the correctness of the implementation are raised

More information

Rank Measures for Ordering

Rank Measures for Ordering Rank Measures for Ordering Jin Huang and Charles X. Ling Department of Computer Science The University of Western Ontario London, Ontario, Canada N6A 5B7 email: fjhuang33, clingg@csd.uwo.ca Abstract. Many

More information

A Box-Consistency Contraction Operator Based on extremal Functions

A Box-Consistency Contraction Operator Based on extremal Functions A Box-Consistency Contraction Operator Based on extremal Functions Gilles Trombettoni, Yves Papegay, Gilles Chabert, Odile Pourtallier INRIA, Université de Nice-Sophia, France October 2008, El Paso, Texas

More information

Bulk Synchronous Parallel ML: Modular Implementation and Performance Prediction

Bulk Synchronous Parallel ML: Modular Implementation and Performance Prediction Bulk Synchronous Parallel ML: Modular Implementation and Performance Prediction Frédéric Loulergue, Frédéric Gava, and David Billiet Laboratory of Algorithms, Complexity and Logic University Paris XII

More information

MINLP Problems and Explanation-based Constraint Programming

MINLP Problems and Explanation-based Constraint Programming MINLP Problems and Explanation-based Constraint Programming Guillaume Rochart 1,2, Eric Monfroy 1, and Narendra Jussien 2 1 LINA, FRE CNRS 2729 2, rue de la Houssinière B.P. 92208 F-44322 Nantes Cedex

More information

HARNESSING CERTAINTY TO SPEED TASK-ALLOCATION ALGORITHMS FOR MULTI-ROBOT SYSTEMS

HARNESSING CERTAINTY TO SPEED TASK-ALLOCATION ALGORITHMS FOR MULTI-ROBOT SYSTEMS HARNESSING CERTAINTY TO SPEED TASK-ALLOCATION ALGORITHMS FOR MULTI-ROBOT SYSTEMS An Undergraduate Research Scholars Thesis by DENISE IRVIN Submitted to the Undergraduate Research Scholars program at Texas

More information

Fast algorithms for max independent set

Fast algorithms for max independent set Fast algorithms for max independent set N. Bourgeois 1 B. Escoffier 1 V. Th. Paschos 1 J.M.M. van Rooij 2 1 LAMSADE, CNRS and Université Paris-Dauphine, France {bourgeois,escoffier,paschos}@lamsade.dauphine.fr

More information

B553 Lecture 12: Global Optimization

B553 Lecture 12: Global Optimization B553 Lecture 12: Global Optimization Kris Hauser February 20, 2012 Most of the techniques we have examined in prior lectures only deal with local optimization, so that we can only guarantee convergence

More information

Today. Golden section, discussion of error Newton s method. Newton s method, steepest descent, conjugate gradient

Today. Golden section, discussion of error Newton s method. Newton s method, steepest descent, conjugate gradient Optimization Last time Root finding: definition, motivation Algorithms: Bisection, false position, secant, Newton-Raphson Convergence & tradeoffs Example applications of Newton s method Root finding in

More information

Discrete Optimization. Lecture Notes 2

Discrete Optimization. Lecture Notes 2 Discrete Optimization. Lecture Notes 2 Disjunctive Constraints Defining variables and formulating linear constraints can be straightforward or more sophisticated, depending on the problem structure. The

More information

A strong local consistency for constraint satisfaction

A strong local consistency for constraint satisfaction A strong local consistency for constraint satisfaction Romuald Debruyne Ecole des Mines de Nantes 4, rue Alfred Kastler, La Chantrerie 4437 Nantes cedex 3 - France Email: debruyne@lirmm.fr Abstract Filtering

More information

Tabu search and genetic algorithms: a comparative study between pure and hybrid agents in an A-teams approach

Tabu search and genetic algorithms: a comparative study between pure and hybrid agents in an A-teams approach Tabu search and genetic algorithms: a comparative study between pure and hybrid agents in an A-teams approach Carlos A. S. Passos (CenPRA) carlos.passos@cenpra.gov.br Daniel M. Aquino (UNICAMP, PIBIC/CNPq)

More information

Efficient Barrier Synchronization Mechanism for the BSP Model on Message-Passing Architectures

Efficient Barrier Synchronization Mechanism for the BSP Model on Message-Passing Architectures Efficient Barrier Synchronization Mechanism for the BSP Model on Message-Passing Architectures Jin-Soo Kim Soonhoi Ha Chu Shik Jhon Department of Computer Engineering Seoul National University Seoul 151-742,

More information

On Computing Minimum Size Prime Implicants

On Computing Minimum Size Prime Implicants On Computing Minimum Size Prime Implicants João P. Marques Silva Cadence European Laboratories / IST-INESC Lisbon, Portugal jpms@inesc.pt Abstract In this paper we describe a new model and algorithm for

More information

ON WEIGHTED RECTANGLE PACKING WITH LARGE RESOURCES*

ON WEIGHTED RECTANGLE PACKING WITH LARGE RESOURCES* ON WEIGHTED RECTANGLE PACKING WITH LARGE RESOURCES* Aleksei V. Fishkin, 1 Olga Gerber, 1 Klaus Jansen 1 1 University of Kiel Olshausenstr. 40, 24118 Kiel, Germany {avf,oge,kj}@informatik.uni-kiel.de Abstract

More information

B.H.GARDI COLLEGE OF ENGINEERING & TECHNOLOGY (MCA Dept.) Parallel Database Database Management System - 2

B.H.GARDI COLLEGE OF ENGINEERING & TECHNOLOGY (MCA Dept.) Parallel Database Database Management System - 2 Introduction :- Today single CPU based architecture is not capable enough for the modern database that are required to handle more demanding and complex requirements of the users, for example, high performance,

More information

Layer-Based Scheduling Algorithms for Multiprocessor-Tasks with Precedence Constraints

Layer-Based Scheduling Algorithms for Multiprocessor-Tasks with Precedence Constraints Layer-Based Scheduling Algorithms for Multiprocessor-Tasks with Precedence Constraints Jörg Dümmler, Raphael Kunis, and Gudula Rünger Chemnitz University of Technology, Department of Computer Science,

More information

An Extension of Complexity Bounds and Dynamic Heuristics for Tree-Decompositions of CSP

An Extension of Complexity Bounds and Dynamic Heuristics for Tree-Decompositions of CSP An Extension of Complexity Bounds and Dynamic Heuristics for Tree-Decompositions of CSP Philippe Jégou, Samba Ndojh Ndiaye, and Cyril Terrioux LSIS - UMR CNRS 6168 Université Paul Cézanne (Aix-Marseille

More information

SPERNER S LEMMA MOOR XU

SPERNER S LEMMA MOOR XU SPERNER S LEMMA MOOR XU Abstract. Is it possible to dissect a square into an odd number of triangles of equal area? This question was first answered by Paul Monsky in 970, and the solution requires elements

More information

Parallel Computing. Slides credit: M. Quinn book (chapter 3 slides), A Grama book (chapter 3 slides)

Parallel Computing. Slides credit: M. Quinn book (chapter 3 slides), A Grama book (chapter 3 slides) Parallel Computing 2012 Slides credit: M. Quinn book (chapter 3 slides), A Grama book (chapter 3 slides) Parallel Algorithm Design Outline Computational Model Design Methodology Partitioning Communication

More information

Parallel Implementation of Interval Analysis for Equations Solving

Parallel Implementation of Interval Analysis for Equations Solving Parallel Implementation of Interval Analysis for Equations Solving Yves Papegay, David Daney, and Jean-Pierre Merlet INRIA Sophia Antipolis COPRIN Team, 2004 route des Lucioles, F-06902 Sophia Antipolis,

More information

New Upper Bounds for MAX-2-SAT and MAX-2-CSP w.r.t. the Average Variable Degree

New Upper Bounds for MAX-2-SAT and MAX-2-CSP w.r.t. the Average Variable Degree New Upper Bounds for MAX-2-SAT and MAX-2-CSP w.r.t. the Average Variable Degree Alexander Golovnev St. Petersburg University of the Russian Academy of Sciences, St. Petersburg, Russia alex.golovnev@gmail.com

More information

Chapter S:II. II. Search Space Representation

Chapter S:II. II. Search Space Representation Chapter S:II II. Search Space Representation Systematic Search Encoding of Problems State-Space Representation Problem-Reduction Representation Choosing a Representation S:II-1 Search Space Representation

More information

Sparse Hypercube 3-Spanners

Sparse Hypercube 3-Spanners Sparse Hypercube 3-Spanners W. Duckworth and M. Zito Department of Mathematics and Statistics, University of Melbourne, Parkville, Victoria 3052, Australia Department of Computer Science, University of

More information

Framework for Design of Dynamic Programming Algorithms

Framework for Design of Dynamic Programming Algorithms CSE 441T/541T Advanced Algorithms September 22, 2010 Framework for Design of Dynamic Programming Algorithms Dynamic programming algorithms for combinatorial optimization generalize the strategy we studied

More information

Module 1 Lecture Notes 2. Optimization Problem and Model Formulation

Module 1 Lecture Notes 2. Optimization Problem and Model Formulation Optimization Methods: Introduction and Basic concepts 1 Module 1 Lecture Notes 2 Optimization Problem and Model Formulation Introduction In the previous lecture we studied the evolution of optimization

More information

Evaluation of Parallel Programs by Measurement of Its Granularity

Evaluation of Parallel Programs by Measurement of Its Granularity Evaluation of Parallel Programs by Measurement of Its Granularity Jan Kwiatkowski Computer Science Department, Wroclaw University of Technology 50-370 Wroclaw, Wybrzeze Wyspianskiego 27, Poland kwiatkowski@ci-1.ci.pwr.wroc.pl

More information

A Fast and Simple Algorithm for Bounds Consistency of the AllDifferent Constraint

A Fast and Simple Algorithm for Bounds Consistency of the AllDifferent Constraint A Fast and Simple Algorithm for Bounds Consistency of the AllDifferent Constraint Alejandro Lopez-Ortiz 1, Claude-Guy Quimper 1, John Tromp 2, Peter van Beek 1 1 School of Computer Science 2 C WI University

More information

Joint Entity Resolution

Joint Entity Resolution Joint Entity Resolution Steven Euijong Whang, Hector Garcia-Molina Computer Science Department, Stanford University 353 Serra Mall, Stanford, CA 94305, USA {swhang, hector}@cs.stanford.edu No Institute

More information

Algorithms for Grid Graphs in the MapReduce Model

Algorithms for Grid Graphs in the MapReduce Model University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln Computer Science and Engineering: Theses, Dissertations, and Student Research Computer Science and Engineering, Department

More information

This article was originally published in a journal published by Elsevier, and the attached copy is provided by Elsevier for the author s benefit and for the benefit of the author s institution, for non-commercial

More information

AUTOMATED REASONING. Agostino Dovier. Udine, October 1, Università di Udine CLPLAB

AUTOMATED REASONING. Agostino Dovier. Udine, October 1, Università di Udine CLPLAB AUTOMATED REASONING Agostino Dovier Università di Udine CLPLAB Udine, October 1, 2018 AGOSTINO DOVIER (CLPLAB) AUTOMATED REASONING UDINE, OCTOBER 1, 2018 1 / 28 COURSE PLACEMENT International Master Degree

More information

Contents. Preface xvii Acknowledgments. CHAPTER 1 Introduction to Parallel Computing 1. CHAPTER 2 Parallel Programming Platforms 11

Contents. Preface xvii Acknowledgments. CHAPTER 1 Introduction to Parallel Computing 1. CHAPTER 2 Parallel Programming Platforms 11 Preface xvii Acknowledgments xix CHAPTER 1 Introduction to Parallel Computing 1 1.1 Motivating Parallelism 2 1.1.1 The Computational Power Argument from Transistors to FLOPS 2 1.1.2 The Memory/Disk Speed

More information

Selecting Topics for Web Resource Discovery: Efficiency Issues in a Database Approach +

Selecting Topics for Web Resource Discovery: Efficiency Issues in a Database Approach + Selecting Topics for Web Resource Discovery: Efficiency Issues in a Database Approach + Abdullah Al-Hamdani, Gultekin Ozsoyoglu Electrical Engineering and Computer Science Dept, Case Western Reserve University,

More information

Floating-Point Numbers in Digital Computers

Floating-Point Numbers in Digital Computers POLYTECHNIC UNIVERSITY Department of Computer and Information Science Floating-Point Numbers in Digital Computers K. Ming Leung Abstract: We explain how floating-point numbers are represented and stored

More information

Floating-Point Numbers in Digital Computers

Floating-Point Numbers in Digital Computers POLYTECHNIC UNIVERSITY Department of Computer and Information Science Floating-Point Numbers in Digital Computers K. Ming Leung Abstract: We explain how floating-point numbers are represented and stored

More information

HPC Algorithms and Applications

HPC Algorithms and Applications HPC Algorithms and Applications Dwarf #5 Structured Grids Michael Bader Winter 2012/2013 Dwarf #5 Structured Grids, Winter 2012/2013 1 Dwarf #5 Structured Grids 1. dense linear algebra 2. sparse linear

More information

Neural Networks. CE-725: Statistical Pattern Recognition Sharif University of Technology Spring Soleymani

Neural Networks. CE-725: Statistical Pattern Recognition Sharif University of Technology Spring Soleymani Neural Networks CE-725: Statistical Pattern Recognition Sharif University of Technology Spring 2013 Soleymani Outline Biological and artificial neural networks Feed-forward neural networks Single layer

More information

Full CNF Encoding: The Counting Constraints Case

Full CNF Encoding: The Counting Constraints Case Full CNF Encoding: The Counting Constraints Case Olivier Bailleux 1 and Yacine Boufkhad 2 1 LERSIA, Université de Bourgogne Avenue Alain Savary, BP 47870 21078 Dijon Cedex olivier.bailleux@u-bourgogne.fr

More information

Reproducibility in Stochastic Simulation

Reproducibility in Stochastic Simulation Reproducibility in Stochastic Simulation Prof. Michael Mascagni Department of Computer Science Department of Mathematics Department of Scientific Computing Graduate Program in Molecular Biophysics Florida

More information

A NUMA Aware Scheduler for a Parallel Sparse Direct Solver

A NUMA Aware Scheduler for a Parallel Sparse Direct Solver Author manuscript, published in "N/P" A NUMA Aware Scheduler for a Parallel Sparse Direct Solver Mathieu Faverge a, Pierre Ramet a a INRIA Bordeaux - Sud-Ouest & LaBRI, ScAlApplix project, Université Bordeaux

More information

Complexity and approximation of satisfactory partition problems

Complexity and approximation of satisfactory partition problems Complexity and approximation of satisfactory partition problems Cristina Bazgan, Zsolt Tuza, and Daniel Vanderpooten LAMSADE, Université Paris-Dauphine, France {bazgan,vdp}@lamsade.dauphine.fr Computer

More information

Scan Scheduling Specification and Analysis

Scan Scheduling Specification and Analysis Scan Scheduling Specification and Analysis Bruno Dutertre System Design Laboratory SRI International Menlo Park, CA 94025 May 24, 2000 This work was partially funded by DARPA/AFRL under BAE System subcontract

More information

Parallel Auction Algorithm for Linear Assignment Problem

Parallel Auction Algorithm for Linear Assignment Problem Parallel Auction Algorithm for Linear Assignment Problem Xin Jin 1 Introduction The (linear) assignment problem is one of classic combinatorial optimization problems, first appearing in the studies on

More information

GraphBLAS Mathematics - Provisional Release 1.0 -

GraphBLAS Mathematics - Provisional Release 1.0 - GraphBLAS Mathematics - Provisional Release 1.0 - Jeremy Kepner Generated on April 26, 2017 Contents 1 Introduction: Graphs as Matrices........................... 1 1.1 Adjacency Matrix: Undirected Graphs,

More information

Delay Estimation for Technology Independent Synthesis

Delay Estimation for Technology Independent Synthesis Delay Estimation for Technology Independent Synthesis Yutaka TAMIYA FUJITSU LABORATORIES LTD. 4-1-1 Kamikodanaka, Nakahara-ku, Kawasaki, JAPAN, 211-88 Tel: +81-44-754-2663 Fax: +81-44-754-2664 E-mail:

More information

Towards ParadisEO-MO-GPU: a Framework for GPU-based Local Search Metaheuristics

Towards ParadisEO-MO-GPU: a Framework for GPU-based Local Search Metaheuristics Towards ParadisEO-MO-GPU: a Framework for GPU-based Local Search Metaheuristics N. Melab, T-V. Luong, K. Boufaras and E-G. Talbi Dolphin Project INRIA Lille Nord Europe - LIFL/CNRS UMR 8022 - Université

More information

Efficient Incremental Mining of Top-K Frequent Closed Itemsets

Efficient Incremental Mining of Top-K Frequent Closed Itemsets Efficient Incremental Mining of Top- Frequent Closed Itemsets Andrea Pietracaprina and Fabio Vandin Dipartimento di Ingegneria dell Informazione, Università di Padova, Via Gradenigo 6/B, 35131, Padova,

More information

Propagate the Right Thing: How Preferences Can Speed-Up Constraint Solving

Propagate the Right Thing: How Preferences Can Speed-Up Constraint Solving Propagate the Right Thing: How Preferences Can Speed-Up Constraint Solving Christian Bessiere Anais Fabre* LIRMM-CNRS (UMR 5506) 161, rue Ada F-34392 Montpellier Cedex 5 (bessiere,fabre}@lirmm.fr Ulrich

More information

Parallel Hybrid Monte Carlo Algorithms for Matrix Computations

Parallel Hybrid Monte Carlo Algorithms for Matrix Computations Parallel Hybrid Monte Carlo Algorithms for Matrix Computations V. Alexandrov 1, E. Atanassov 2, I. Dimov 2, S.Branford 1, A. Thandavan 1 and C. Weihrauch 1 1 Department of Computer Science, University

More information

Real Numbers finite subset real numbers floating point numbers Scientific Notation fixed point numbers

Real Numbers finite subset real numbers floating point numbers Scientific Notation fixed point numbers Real Numbers We have been studying integer arithmetic up to this point. We have discovered that a standard computer can represent a finite subset of the infinite set of integers. The range is determined

More information

Why AC-3 is Almost Always Better than AC-4 for Establishing Arc Consistency in CSPs*

Why AC-3 is Almost Always Better than AC-4 for Establishing Arc Consistency in CSPs* Why AC-3 is Almost Always Better than AC-4 for Establishing Arc Consistency in CSPs* Richard J. Wallace Department of Computer Science, University of New Hampshire Durham, NH 03824, U. S. A. Abstract On

More information

Search techniques for non-linear constraint satisfaction problems with inequalities

Search techniques for non-linear constraint satisfaction problems with inequalities Search techniques for non-linear constraint satisfaction problems with inequalities Marius-C alin Silaghi, Djamila Sam-Haroud, and Boi Faltings Arti cial Intelligence Laboratory Swiss Federal Institute

More information

IMPLEMENTATION OF A FIXING STRATEGY AND PARALLELIZATION IN A RECENT GLOBAL OPTIMIZATION METHOD

IMPLEMENTATION OF A FIXING STRATEGY AND PARALLELIZATION IN A RECENT GLOBAL OPTIMIZATION METHOD IMPLEMENTATION OF A FIXING STRATEGY AND PARALLELIZATION IN A RECENT GLOBAL OPTIMIZATION METHOD Figen Öztoprak, Ş.İlker Birbil Sabancı University Istanbul, Turkey figen@su.sabanciuniv.edu, sibirbil@sabanciuniv.edu

More information

Parameterized Algorithm for Eternal Vertex Cover

Parameterized Algorithm for Eternal Vertex Cover Parameterized Algorithm for Eternal Vertex Cover Fedor V. Fomin a,1, Serge Gaspers b, Petr A. Golovach c, Dieter Kratsch d, Saket Saurabh e, a Department of Informatics, University of Bergen, N-5020 Bergen,

More information

Computational Complexity of Multi-way, Dataflow Constraint Problems

Computational Complexity of Multi-way, Dataflow Constraint Problems Computational Complexity of Multi-way, Dataflow Constraint Problems Gilles Trombettoni and Bertrand Neveu Projet Contraintes, CERMICS/INRIA, 2004 route des lucioles, 06902 Sophia-Antipolis Cedex, B.P.

More information

Fast Fuzzy Clustering of Infrared Images. 2. brfcm

Fast Fuzzy Clustering of Infrared Images. 2. brfcm Fast Fuzzy Clustering of Infrared Images Steven Eschrich, Jingwei Ke, Lawrence O. Hall and Dmitry B. Goldgof Department of Computer Science and Engineering, ENB 118 University of South Florida 4202 E.

More information

Clustering. Informal goal. General types of clustering. Applications: Clustering in information search and analysis. Example applications in search

Clustering. Informal goal. General types of clustering. Applications: Clustering in information search and analysis. Example applications in search Informal goal Clustering Given set of objects and measure of similarity between them, group similar objects together What mean by similar? What is good grouping? Computation time / quality tradeoff 1 2

More information

An Extension of the WAM for Hybrid Interval Solvers

An Extension of the WAM for Hybrid Interval Solvers Appears in: Journal of Functional and Logic Programming. Special issue on Parallelism and Implementation Technology for Constraint Logic Programming. MIT Press 1999 An Extension of the WAM for Hybrid Interval

More information

Efficient validation and construction of border arrays

Efficient validation and construction of border arrays Efficient validation and construction of border arrays Jean-Pierre Duval Thierry Lecroq Arnaud Lefebvre LITIS, University of Rouen, France, {Jean-Pierre.Duval,Thierry.Lecroq,Arnaud.Lefebvre}@univ-rouen.fr

More information

The Encoding Complexity of Network Coding

The Encoding Complexity of Network Coding The Encoding Complexity of Network Coding Michael Langberg Alexander Sprintson Jehoshua Bruck California Institute of Technology Email: mikel,spalex,bruck @caltech.edu Abstract In the multicast network

More information

Flight Systems are Cyber-Physical Systems

Flight Systems are Cyber-Physical Systems Flight Systems are Cyber-Physical Systems Dr. Christopher Landauer Software Systems Analysis Department The Aerospace Corporation Computer Science Division / Software Engineering Subdivision 08 November

More information