Italian FIRB Project on
 Parallel Algorithms and Numerical Nonlinear Optimization

Conference Communications

International Conferences

  1. Ya.D. Sergeyev, Infinity Computer and Calculus (invited plenary lecture), in Proc. of the 3rd International Conference of Applied Mathematics TICAM, Plovdiv (Bulgaria), August 12-18, 2006, Bainov D. and Nenov S., eds., University of Plovdiv Press, Vol. 2, pp. 246-247.

    2. [Top]

  2. Ya.D. Sergeyev, Infinity Computer and Calculus (invited plenary lecture), in Proc. of the Conference on Applied Optimization and Metaheuristic Innovations, Yalta (Ukraine), July 19-21, 2006, Sergienko I. V., ed., pp. 40-41.

    3. [Top]

  3. G. Zanghirati, L. Zanni, Some properties of gradient-based methods with application to machine learning (invited talk), EuroXXI - 21st European Conference on Operational Research, Reykjavik (Iceland), July 2-5, 2006.

    4. [Top]

  4. G. Zanghirati, L. Zanni, Large-scale Support Vector Machines: Decomposition and Cascade Approaches (invited talk) EuroXXI - 21st European Conference on Operational Research, Reykjavik (Iceland), July 2-5, 2006.

    5. [Top]

  5. G. Spaletta, Modelling the Thyroid Geometry, 8th International Mathematica Symposium, Palais des Papes, Avignone (France), June 19-23, 2006.

    6. [Top]

  6. M. Sofroniou, Mathematica's Numerics (series of invited talks), 8th International Mathematica Symposium, Palais des Papes, Avignone (France), June 19-23, 2006.

    7. [Top]

  7. G. Spaletta, M. Sofroniou, Matrix Polynomials and the Matrix Exponential (invited talk), Castellon Conference on Geometric Integration, University Jaume I, September 18-22, 2006.

    8. [Top]

  8. E. Galligani, V. Ruggiero, G. Zanghirati, Splitting methods for nonlinear diffusion filtering, 3rd IASC World Conference on "Computational Statistics and Data Analysis", Limassol (Cyprus), October 28-31, 2005.

    9. [Top]

  9. T. Serafini, G. Zanghirati, L. Zanni, Gradient Projection-Type Quadratic Solvers in Parallel Decomposition Techniques for Support Vector Machines, 3rd IASC World Conference on "Computational Statistics and Data Analysis", Limassol (Cyprus), October 28-31, 2005.

    10. [Top]

  10. T. Serafini, G. Zanghirati, L. Zanni, Some Improvements to a Parallel Decomposition Technique for Training Support Vector Machines, EURO PVM MPI 2005, 12th European Parallel Virtual Machine and Message Passing Interface Conference, Sorrento (Naples, Italy), September 18-21, 2005.

    11. [Top]

  11. T. Serafini, G. Zanghirati, L. Zanni, Decomposition techniques and gradient projection methods in training Support Vector Machines, SIAM Conference on Optimization 2005, Stocholm (Sweden), May 15-19, 2005.

    12. [Top]

  12. T. Serafini, G. Zanghirati, L. Zanni, On gradient projection-based decomposition techniques for training SVMs on parallel architectures, PASCAL Workshop, Thurnau (Germany), March 16-18, 2005.

    13. [Top]

  13. Ya.D. Sergeyev, D.E. Kvasov, Diagonal global search based on a set of possible Lipschitz constants, In Proc. of the International Workshop on Global Optimization GO05, Almerìa (Spain), September 18-22, 2005, pp. 219-224.

    14. [Top]

  14. YA.D.Sergeyev, Global search based on efficient diagonal partitions (invited plenary lecture), International Conference on Complementarity, Duality, and Global Optimization in Science and Engineering CDGO2005, Blacksburg (Virginia, USA), August 15-17, 2005. .

    15. [Top]

  15. V.A. Grishagin, Ya.D. Sergeyev, Parallel algorithms for multidimensional global optimization with non-convex constraints, in Proc. of the 5th International Workshop "High-Performance Computing on Clusters", Nizhni Novgorod "Lobachevsky" University, Nizhni Novgorod (Russia) November 22-25, 2005, R.G. Strongin, Ed., NNGU Press, pp. 74-83.

    16. [Top]

  16. F.Zama, Image Processing in FEMLAB using Diffusion Filters, Femlab Conference 2005, Stocholm (Sweden), October 3-5, 2005.

    17. [Top]

  17. Ya.D. Sergeyev, Infinity Computer: Principles of work and applications (invited plenary lecture), in Proc. of the International Conference on Selected Problems of Modern Mathematics, Ishanov S.I., Ed., Kaliningrad (Russia), April 4-8, 2005, pp. 211-213.

    18. [Top]

  18. G. Landi, The Lagrange Method for the Regularization of Discrete Ill-Posed Problems, 22nd International Federation for Information Processing Conference on System Modeling and Optimization TC 7 Conference, Torino, 18-22 Luglio 2005.

    19. [Top]

  19. V.A. Grishagin, Ya.D. Sergeyev (2004), Efficiency of parallelization of characteristic global optimization algorithms in the framework of the nested optimization scheme, Proceedings of the 4th International Workshop "Parallel Computations on Clusters", Samara, Russia, September 30 - October 2, 2004, Soyfer V. ed., RC RAN Press, pp. 70-74.

    20. [Top]

  20. Ya.D. Sergeyev (2004), A new computational paradigm: Mathematical model and applications, Proceedings of the  VI-th International Congress on "Mathematical Modeling", Nizhni Novgorod State University, Nizhni Novgorod, Russia, September 20-26, p. 27.

    21. [Top]

  21. Ya.D. Sergeyev, D. E. Kvasov (2004), Lipschitzian global optimization without the Lipschitz constant based on adaptive diagonal curves, Proceedings of the VI-th International Congress on "Mathematical Modeling", Nizhni Novgorod State University, Nizhni Novgorod, Russia, September 20-26, p. 121.

    22. [Top]

  22. M. Sofroniou, G. Spaletta (invited), The Matrix Exponential: Efficient Computation and Error Analysis, SIMAI XIV, September 20-24, 2004, Venice (Italy).

    23. [Top]

  23. G. Spaletta (invited), Numerical Assessments in the Work of Vito Volterra, ICMMB XIV, September 16-18, 2004, Bologna (Italy).

    24. [Top]

  24. E. Loli Piccolomini, A descent method for computing the Tikhonov regularized solution of linear inverse problems, SPIE Annual Meeting "Optical Science and Technology", Denver, Colorado (USA), August 2-6, 2004.

    25. [Top]

  25. G. Landi, A Total Variation based Regularization strategy in Magnetic Resonance Imaging, SPIE Annual Meeting "Optical Science and Technology", Denver, Colorado (USA), August 2-6, 2004.

    26. [Top]

  26. Y. Sergeyev, Global optimization methods and classes of test functions, First International Conference on Continuous Optimization ICCOPT-I. Troy (NY), USA, August 2-4, 2004.

    27. [Top]

  27. Y. Sergeyev, Algorithms and partition strategies for Lipschitzian global optimization, 40th Workshop Large Scale Nonlinear Optimization, Erice (TP), Italy, June 22 - July 1, 2004.

    28. [Top]

  28. T. Serafini, G. Zanghirati, L. Zanni, Recent improvements to gradient projection-based decomposition techniques for Support Vector Machines, International Conference "MML2004 - Mathematical Methods for Learning", June 21-24, 2004, Como (Italy).

    29. [Top]

  29. G. Spaletta, M. Sofroniou, Solving Linear Systems Accurately, Workshop on Dynamical Systems on Matrix Manifolds, May 27-28, 2004, Bari (Italy).

    30. [Top]

  30. S. Bonettini, V. Ruggiero, E. Galligani, Some iterative methods for the solution of a reduced symmetric indefnite KKT system, International Conference "HPSNO2004 - High Performance Algorithms and Software for Nonlinear Optimization", May 18-20, 2004, Ischia (Naples, Italy).

    31. [Top]

  31. T. Serafini, G. Zanghirati, L. Zanni, A Gradient Projection-based Decomposition Software for Large Quadratic Programs in Training Support Vector Machines, International Conference "HPSNO2004 - High Performance Algorithms and Software for Nonlinear Optimization", May 18-20, 2004, Ischia (Naples, Italy).

    32. [Top]

  32. S. Bonettini, V. Ruggiero, E. Galligani, Interior point method as inexact Newton method for KKT systems CORS/INFORMS Joint Int. Meeting, May 16-19, 2004, Banff, Alberta (Canada).

    33. [Top]

  33. G. Spaletta, M. Sofroniou, Efficient Matrix Polynomial Computation and
    Application to the Matrix Exponential    , IWTAM II, April 1-3, 2004, Montecatini (Italy).

    34. [Top]

  34. V. Ruggiero (invited), Interior point method as inexact Newton method for KKt systems, Second International Workshop on the Technological Aspects of Mathematics, April 1-3, 2004, Montecatini (Italy).

    35. [Top]

Communications at the Scientific Meeting "Numerical Methods for Local and Global Optimization: Sequential and Parallel Algorithms", INdAM Conference in Cortona (Italy), July 14-20, 2003.

  1. G. Landi, E. Loli Piccolomini (2003), The Total Variation Regularization in Dynamic MR Imaging.
    Abstract.  The Total Variation (TV) regularization method, proposed by Rudin, Osher and Fatemi in [Physica D., 1992], has recently become a very popular technique in image restoration problems. The motivation for its success is that the TV regularization performs well for denoising and deblurring while preserving sharp discontinuities. The good performance of the TV model makes it particularly attractive for biomedical imaging. We study the use of TV regularization technique in the reconstruction of dynamic Magnetic Resonance images [Z.P. Liang, P.C. Lauterbur, 1994]. Several iterative schemes have been proposed in the literature for solving the TV minimization problem. We compare a Newton and a fixed point method [C.R. Vogel, M.E. Oman, 1996] performances on both test problems and real MR data.

    2. [Top]

  2. F. Zama, E. Loli Piccolomini, Truncated Conjugate Gradient Iterations for Solving Ill-Posed Problems.
    Abstract.  A large variety of applications give raise naturally to ill-posed problems. Whenever the underlying physical or technical problem is modelled by an integral equation of the first kind with a smooth kernel. The data usually stem from measurements with a limited precision, i.e., only perturbed data are available. The inverse problem is ill-posed and requires regularization methods. In this work we describe an iterative algorithm for finding the solution and the regularization parameter. Truncated Conjugate Gradients Iterations are implemented for computing the solution, while the value of the regularization is determined in order to have decreasing values of the objective functional. We develop a stopping criterion for the CG-iterates which is linked to the noise level and the current value of the regularization parameter.

    3. [Top]

  3. G. Spaletta, M. Sofroniou, Symmetric Composition of Symmetric Numerical Integration Methods.
    Abstract.  This work focuses on the derivation of composition methods for the numerical integration of ordinary differential equations. In contrast to the Aitken-Neville algorithm used in extrapolation, composition can conserve desirable geometric properties of a base integration method, such as symplecticity [E. Hairer, Ch. Lubich, G. Wanner, 2002]. We survey existing composition methods and describe results of a numerical search for new methods [R.I. McLachlan, G.R.W. Quispel, 2002; H. Yoshida, 1990]. The optimization procedure that has been adopted [P.E. Gill, W. Murray, M.H. Wright, 1984] can be very intensive, especially when deriving high order composition schemes. To overcome this, we make use of parallel computation. Numerical examples indicate that the new methods perform better than previously known methods.

    4. [Top]

  4. F. Tinti, Numerical Solution of Pseudomonotone Variational Inequalities Problems by Extragradient Methods.
    Abstract.  In this work we have analyzed from the numerical point of view the class of projection methods for solving variational inequality problems. We focus on some specific extragradient methods making use of different choices of the steplengths. Subsequently we have analyzed the hyperplane projection methods in which we construct an appropriate hyperplane which strictly separates the current iterate from the solutions of the problem. Finally we have included a collection of test problems implemented in Matlab to illustrate the effectiveness of the proposed methods.

    5. [Top]

  5. S. Bonettini, A Nonmonotone Inexact Newton Method.
    Abstract.  In this work we describe a variant of the inexact Newton method for solving nonlinear systems of equations. We define a nonmonotone inexact Newton step and a nonmonotone backtracking strategy. For this nonmonotone scheme we present the convergence theorems. Finally, we show how we can apply these strategies to Newton inexact interior-point method and we present some numerical examples.

    6. [Top]

  6. T. Serafini, G. Zanghirati, L. Zanni, Parallel Training of Support Vector Machines.
    Abstract.  We present a parallel approach for the solution of quadratic programming (QP) problems with box constraints and a single linear constraint, arising in the training of Support Vector Machines. In this kind of application the problem is dense and generally large-sized (larger then 104). An iterative decomposition technique has been presented in [T. Joachims, 1998], which solves the problem by splitting it into a sequence of very small QP (inner) subproblems (generally with size less than 102). The approach proposed in [G. Zanghirati, L. Zanni, 2003] follows this decomposition idea, but it is designed to split the whole problem into QP subproblems of sufficiently large size (> 103), so that they can be efficiently solved by special gradient projection methods [T. Serafini, G. Zanghirati, L. Zanni, 2003]. On scalar architectures this new technique allows for comparable performance with those of the algorithm in [T. Joachims, 1998], but it is much more suited for parallel implementations. In fact, parallel versions of the gradient projection methods can be applied to solve the large QP inner subproblems and the other expensive tasks of each decomposition iteration can be easily distributed among the available processors. We present several improvements of this approach and we evaluate their effectiveness on large-scale benchmark problems both in scalar and parallel environments.

    7. [Top]

  7. M. Gaviano, Complexity analysis in miniimization problems.
    Abstract.  The investigation of the numerical complexity of algorithms that minimize functions from Rn into R is a very difficult issue. In recent years new results have been found giving useful guidelines for a better understanding of the algorithm behaviors and even for improving their performances. Many results established for general or specific minimization problems are reviewed. The case of algorithms for global minimization is also considered.

    8. [Top]

  8. D. Lera, Stochastic Global Optimization Methods.
    Abstract. In this seminar we will present some stochastic techniques for solving global optimization problems. Stochastic methods are techniques that contain some stochastic elements. This means that either the outcome of the method is itself a random variable (i.e. algorithms in which the placement of observations is based on the generation of random points in the domain of the objective function), or we consider the objective function to be a realization of a stochastic process. Excellent surveys on the subject are in [B. Betrò, 1991; C.G.E. Boender, H.E. Romeijn, 1995; F. Schoen 1991; A.A. Torn, A. Zilinskas, 1989]. Here we will discuss, in particular, the so-called two-phase methods, i.e. methods in which both a global phase and a local phase can be distinguished. In the global phase, the function is evaluated in a number of randomly sampled points. During the local phase the sample points are "manipulated" in order to yield a candidate global minimum. We will give a brief presentation of clustering techniques and finally we will show random search methods and Simulated Annealing algorithms.

    9. [Top]

  9. D.E. Kvasov, M. Gaviano, D. Lera, Ya.D. Sergeyev, GKLS-Generator for Testing Global Optimization Algorthms.
    Abstract.  Development of efficient global optimization algorithms is impossible without examination of their behaviour on sets of sophisticated test problems. The lack of complete information (such as number of local minima, their locations, attraction regions, local and global values, ecc.) describing global optimization tests taken from real-life applications does not allow to use themfor studying and verifying validity of the algorithms. That is why a significant effort is made to construct test functions [C.A. Floudas et al, 1994; C.A. Floudas et al., 1999; M. Gaviano, D. Lera, 1998; J. Pintèr, 2002]. In this communication, a procedure for generating three types (non-differentiable, continuously differentiable, and twice continuously differentiable) of classes of multidimensional and multiextremal test functions with known local and global minima is presented. The procedure consists of defining a convex quadratic function systematically distorted by polynomials. Each test class provided by the GKLS-generator consists of 100 functions and is defined by the following parameters: (i) problem dimension, (ii) number of local minima, (iii) global minimum value, (iv) radius of the attraction region of the global minimizer, (v) distance from the global minimizer to the quadratic function vertex. The other necessary parameters are chosen randomly by the generator for each test function of the class. A special notebook with a complete description of all functions is supplied to the user.

    10. [Top]

  10. L. Zanni, Gradient Projection Methods for Quadratic Programs.
    Abstract.  Gradient projection methods for quadratic programming problems with simple constraints are considered. Starting from the analysis of the classical versions of these schemes, some recent gradient projection algorithms are introduced and the importance of using appropriate linesearch strategies and steplength selection rules is stressed. Linesearch techniques based on both limited minimization rules and nonmonotone strategies are considered. For the steplength selection, the Barzilai-Borwein spectral rules are discussed and some new suggestions are presented. Numerical results showing the behaviours of the proposed approaches are reported.

    [Top]

Communications at the 4th International Conference "4th International Conference on Global Optimization", Santonini (Greece), June 8-12, 2003.
  1. Ya.D. Sergeyev, Lipschitz Global Optimization and Local Information, Proc. of the 4th Inter. Conf. on Frontiers in Global Optimization, Aegean Conferences Series 10, p. 21.

    2. [Top]

  2. D.E. Kvasov, M. Gaviano, D. Lera, Ya.D. Sergeyev, Generator of Classes ofTest Functions for Global Optimization Algorthms, Proc. of the 4th Inter. Conf. on Frontiers in Global Optimization, Aegean Conferences Series 10, p. 55.

    3. [Top]

  3. L.G. Casado, J.A. Martinez, I. Garcia, Ya.D. Sergeyev, B. Toth, Efficient Use of Gradient Information in Multidimensional Interval Global Optimization Algorthms, Proc. of the 4th Inter. Conf. on Frontiers in Global Optimization, Aegean Conferences Series 10, p. 61.

    4. [Top]

Other talks at International Conferences

  1. G.Spaletta, M. Sofroniou, Solving orthogonal matrix differential systems in Mathematica, ICCS'02, April 21-24, 2002, Amsterdam (The Netherlands).

    2. [Top]

  2. M. Sofroniou, G. Spaletta, P.C. Moan, G.R.W. Quispel, A generalization of Runge-Kutta methods, CSC 2002, Geneva University, Mathematics Section, June 26-29, 2002, Geneva (Switzerland).

    3. [Top]

  3. T. Serafini, G. Zanghirati, L. Zanni, Adaptive Steplength Selections in Gradient Projection Methods for QP, "NA03 - 20th Biennial Conference on Numerical Analysis", June 24-27, 2003, Dundee (Scothland).

    4. [Top]

  4. T. Serafini, G. Zanghirati, L. Zanni, Parallel Decomposition Approaches for Training SVMs, International Conference "ParCo2003", September 2-5, 2003, Dresda (Germany).

    5. [Top]

  5. G. Zanghirati, L. Zanni, Decomposition Techniques in Training  Support Vector Machines: Inner QP Solvers and Parallel Approaches, International Workshop on "Mathematical Diagnostics", June 17-26, 2002, Erice (Italy).

    6. [Top]

  6. G. Zanghirati, L. Zanni, Variable Projection Methods for Quadratic Programs in Training Support Vector Machines with Gaussian Kernels, International Conference "SIAM Meeting on Optimization 2002", May 20-22, 2002, Toronto (Canada).

    7. [Top]

National Conferences

  1. G.Spaletta, Some invariance theorems for one--step integration methods (invited talk), Structural Dynamical Systems: Computational Aspects (SDS 2006), Monopoli (Bari, Italy), June 13-16, 2006.

    2. [Top]

  2. Ya.D. Sergeyev, Infinity Computer and Calculus, Atti del convegno "SIMAI 2006 - VII Congress of the Italian Society for Applied and Industrial Mathematics", Baia Samuele (Ragusa, Italy), May 22-26, 2006, a cura di Puccio L. et al., Università degli Studi di Messina, 2006, p. 230.

    3. [Top]

  3. G.Spaletta, Concetti di Forma Geometrica ed Integrazione: introduzione alla Approssimazione di Dati Sperimentali, (Lettura Magistrale su invito), Università di Parma, March 29, 2006.

    4. [Top]

  4. T. Serafini, G. Zanghirati, L. Zanni, Numerical Topics on SVMs Classification, workshop "ASTAA Project Meeting 2005", Genova, 9-10 Giugno 2005.

    5. [Top]

  5. V. Ruggiero (invited), Ottimizzazione e Calcolo Parallelo, Workshop honoring Alfonso Laratta, October 13, 2004, Modena (Italy).

    6. [Top]

  6. G. Spaletta (invited), M. Sofroniou, The Matrix Exponential: Efficient Computation and Error Analysis, Congresso SIMAI VII, September 20-24, 2004, Venice (Italy).

    7. [Top]

  7. T. Serafini, G. Zanghirati, L. Zanni, Regole adattative per linesearch e  selezione del passo in metodi del gradiente proiettato  per l’ottimizzazione non lineare, Convegno GNCS, February 9-11, 2004, Montecatini.

    8. [Top]

  8. E. Galligani, V. Ruggiero, S. Bonettini, A Perturbed-Damped Newton Method for Large-Scale Constrained Optimization, National Congress "Analisi Numerica: Stato dell'Arte", September 2002, Rende (Cosenza).

    9. [Top]

  9. S. Bonettini, E. Galligani, V. Ruggiero, Analisi del Metodo di Newton del Punto Interno, XVII Congresso UMI, September 8-13, 2003, Milan.

    10. [Top]

  10. S. Bonettini, E. Galligani, V. Ruggiero, On the Newton Interior-Point Method for Nonlinear Programming, AIRO Conference, September 2-5, 2003, Venice.

    11. [Top]

  11. G. Zanghirati, L. Zanni, A Parallel Solver for Large Quadratic Programs in Training Support Vector Machines, National Conference "SIMAI 2002", May 26-31, 2002, Chia (Cagliari).

    12. [Top]

  12. G. Zanghirati, L. Zanni, Decomposition Techniques for Large Quadratic Programs in Training Support Vector Machines, International Conference "APMOD 2002", June 17-21, 2002, Milan.

    13. [Top]

  13. T. Serafini, G. Zanghirati, L. Zanni, Accelerazione della Convergenza in metodi del Gradiente Proiettato per Problemi di Programmazione Quadratica, XVII Congresso UMI, September 8-13, 2003, Milan.

    14. [Top]

  14. T. Serafini, G. Zanghirati, L. Zanni, Variable Projection Decomposition Techniques for Large-Scale Support Vector Machines, National Congress "Analisi Numerica: Stato dell'Arte", September 2002, Rende (Cosenza).

    15. [Top]

  15. T. Serafini, G. Zanghirati, L. Zanni, Steplength Selections in Gradient Projection Methods for Large-Scale Quadratic Programs, AIRO Conference, September 2-5, 2003, Venice.

    16. [Top]

  16. G. Spaletta, Symplectic Elementary Differential Runge-Kutta Methods, June 22-25, 2003, Monopoli (Bari).

    17. [Top]

  17. Y. Sergeyev, L'infinito in matematica, fisica e filosofia, Pisa, Italy. March 26, 2004.

    18. [Top]

  18. Y. Sergeyev, First Workshop of DEIS, Cetraro (CS), Italy, July 6-8, 2004.

    [Top]

Summer Schools

  1. Metodi Numerici per Equazioni di Evoluzione, Prof. A. Ostermann (Innsbruck University), Prof. J.G. Verwer (CWI, The Netherlands),
    Dobbiaco (Italy), June 28 - July 2, 2004.
    .

    2. [Top]

 

     Top

     

     



Status: completed.
Info: g.zanghirati@unife.it

 Last updated: 15/3/2007.