Results

**11 - 16**of**16**### Recollections on the discovery of the reverse search technique

"... Komei Fukuda and I discovered the idea for reverse search during conversations in Tokyo in October 1990. We were working on the vertex enumeration problem for convex polyhedra. At the time I was visiting Masao Iri at the University of Tokyo and Masakazu Kojima at Tokyo Institute of Technology, suppo ..."

Abstract
- Add to MetaCart

(Show Context)
Komei Fukuda and I discovered the idea for reverse search during conversations in Tokyo in October 1990. We were working on the vertex enumeration problem for convex polyhedra. At the time I was visiting Masao Iri at the University of Tokyo and Masakazu Kojima at Tokyo Institute of Technology, supported by a JSPS/NSERC bilateral exchange. Komei was then working at the University of Tsukuba, Otsuka, which was a couple of subway stations from my office. So we met quite often. One day Komei visited me at my office in Todai and explained to me the criss-cross method for solving linear programs, independently developed by S. Zionts [11], T. Terlaky [8, 9] and Z. Wang [10]. In this method one pivots in the hyperplane arrangement generated by the constraints of the linear program, without regard for feasibility- thus differentiating it from the simplex method. Komei and Tomomi Matsui, a Ph.D. student at Tokyo Institute of Technology, had developed an elegant new proof of the convergence of the criss-cross method [6], which Komei was explaining to me. Komei had drawn a line-arrangement on the blackboard, along with the path the criss-cross method would take from any given vertex to the optimum vertex of the LP. On the board all of these edges were shown in yellow with directions that eventually

### Edmonds Fukuda Rule And A General Recursion For Quadratic Programming

"... A general framework of nite algorithms is presented here for quadratic programming. This algorithm is a direct generalization of Van der Heyden's algorithm for the linear complementarity problem and Jensen's `relaxed recursive algorithm', which was proposed for solution of Oriented Ma ..."

Abstract
- Add to MetaCart

A general framework of nite algorithms is presented here for quadratic programming. This algorithm is a direct generalization of Van der Heyden's algorithm for the linear complementarity problem and Jensen's `relaxed recursive algorithm', which was proposed for solution of Oriented Matroid programming problems. The validity of this algorithm is proved the same way as the finiteness of the criss-cross method is proved. The second part of this paper contains a generalization of Edmonds-Fukuda pivoting rule for quadratic programming. This generalization can be considered as a finite version of Van de Panne - Whinston algorithm and so it is a simplex method for quadratic programming. These algorithms uses general combinatorial type ideas, so the same methods can be applied for oriented matroids as well. The generalization of these methods for oriented matroids is a subject of another paper.

### A NEW ADMISSIBLE PIVOT METHOD FOR LINEAR PROGRAMMING

, 2003

"... We present a new admissible pivot method for linear programming that works with a sequence of improving primal feasible interior points and dual feasible interior points. This method is a practicable variant of the short admissible pivot sequence algorithm, which was suggested by Fukuda and Terlaky. ..."

Abstract
- Add to MetaCart

We present a new admissible pivot method for linear programming that works with a sequence of improving primal feasible interior points and dual feasible interior points. This method is a practicable variant of the short admissible pivot sequence algorithm, which was suggested by Fukuda and Terlaky. Here, we also show that this method can be modified to terminate in finite pivot steps. Finedly, we show that this method outperforms Terlalcy's criss-cross method by computational experiments.

### The Criss-Cross Method Can Take \Omega (nd) Pivots

, 2004

"... Abstract Using deformed products of arrangements, we construct a family of linear programs with ninequalities in! d on which, in the worst-case, the least-index criss-cross method requires \Omega (nd) (for fixed d) pivots to reach optimality. ..."

Abstract
- Add to MetaCart

(Show Context)
Abstract Using deformed products of arrangements, we construct a family of linear programs with ninequalities in! d on which, in the worst-case, the least-index criss-cross method requires \Omega (nd) (for fixed d) pivots to reach optimality.