Hybrid extreme point tabu search

Abstract We develop a new hybrid tabu search method for optimizing a continuous differentiable function over the extreme points of a polyhedron. The method combines extreme point tabu search (EPTS) with traditional descent algorithms based on linear programming. The tabu search algorithm utilizes both recency-based and frequency-based memory and oscillates between local improvement and diversification phases. The hybrid algorithm iterates between using a descent algorithm to find a local minimum and tabu search to improve locally and then move to a new area of the search space. The descent algorithm acts as a form of intensification within the tabu search. This algorithm can be used on many important classes of problems in global optimization including bilinear programming, multilinear programming, multiplicative programming, concave minimization, and complementarity problems. The algorithm is applied to two practical problems: the quasistatie multi-rigid-body contact problem (QCP) in robotics and the global tree optimization (GTO) problem in machine learning. We perform computational experiments on problems of significant real world dimensions: the smallest machine learning problem tested has on the order of 10485 extreme points. The results show our method obtains high quality solutions both by comparison with the embedded descent algorithm and by comparison to a version of tabu search that does not make use of descent.

[1]  Kurt Jörnsten,et al.  Tabu search within a pivot and complement framework , 1994 .

[2]  Kurt Jörnsten,et al.  Tabu Search for General Zero-One Integer Programs Using the Pivot and Complement Heuristic , 1994, INFORMS J. Comput..

[3]  Kristin P. Bennett,et al.  Decision Tree Construction Via Linear Programming , 1992 .

[4]  Nimrod Megiddo,et al.  On the complexity of polyhedral separability , 1988, Discret. Comput. Geom..

[5]  Minghe Sun,et al.  A tabu search heuristic procedure for the fixed charge transportation problem , 1998, Eur. J. Oper. Res..

[6]  Jeffrey C. Trinkle,et al.  Prediction of the quasistatic planar motion of a contacted rigid body , 1995, IEEE Trans. Robotics Autom..

[7]  F. Glover.  Tabu Search Fundamentals and Uses , 1995 .

[8]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[9]  R. Detrano,et al.  International application of a new probability algorithm for the diagnosis of coronary artery disease. , 1989, The American journal of cardiology.

[10]  M. Seetharama Gowda,et al.  The extended linear complementarity problem , 1995, Math. Program..

[11]  Kristin P. Bennett,et al.  Bilinear separation of two sets inn-space , 1993, Comput. Optim. Appl..

[12]  Minghe Sun,et al.  A Tabu Search Heuristic Procedure for Solving the Transportation Problem with Exclusionary Side Constraints , 1998, J. Heuristics.

[13]  Catherine Blake,et al.  UCI Repository of machine learning databases , 1998 .

[14]  Alberto Maria Segre,et al.  Programs for Machine Learning , 1994 .

[15]  Aimo A. Törn,et al.  Global Optimization , 1999, Science.

[16]  P. McKeown.  A branch‐and‐bound algorithm for solving fixed charge problems , 1981 .

[17]  O. Mangasarian,et al.  Multisurface method of pattern separation for medical diagnosis applied to breast cytology. , 1990, Proceedings of the National Academy of Sciences of the United States of America.

[18]  Minghe Sun,et al.  Tabu search applied to the general fixed charge problem , 1993, Ann. Oper. Res..

[19]  Jeffrey C. Trinkle,et al.  A complementarity approach to a quasistatic multi-rigid-body contact problem , 1996, Comput. Optim. Appl..

[20]  Philip Wolfe,et al.  An algorithm for quadratic programming , 1956 .