The nonlinear geometry of linear programming. I. Affine and projective scaling trajectories
Authors:
D. A. Bayer and J. C. Lagarias
Journal:
Trans. Amer. Math. Soc. 314 (1989), 499526
MSC:
Primary 90C05; Secondary 34A99, 52A40
DOI:
https://doi.org/10.1090/S00029947198910055256
MathSciNet review:
1005525
Fulltext PDF Free Access
Abstract  References  Similar Articles  Additional Information
Abstract: This series of papers studies a geometric structure underlying Karmarkar’s projective scaling algorithm for solving linear programming problems. A basic feature of the projective scaling algorithm is a vector field depending on the objective function which is defined on the interior of the polytope of feasible solutions of the linear program. The geometric structure studied is the set of trajectories obtained by integrating this vector field, which we call $P$trajectories. We also study a related vector field, the affine scaling vector field, and its associated trajectories, called $A$trajectories. The affine scaling vector field is associated to another linear programming algorithm, the affine scaling algorithm. Affine and projective scaling vector fields are each defined for linear programs of a special form, called strict standard form and canonical form, respectively. This paper derives basic properties of $P$trajectories and $A$trajectories. It reviews the projective and affine scaling algorithms, defines the projective and affine scaling vector fields, and gives differential equations for $P$trajectories and $A$trajectories. It shows that projective transformations map $P$trajectories into $P$trajectories. It presents Karmarkar’s interpretation of $A$trajectories as steepest descent paths of the objective function $\left \langle {{\mathbf {c}},{\mathbf {x}}} \right \rangle$ with respect to the Riemannian geometry $d{s^2} = \sum \nolimits _{i = 1}^n {d{x_i}d{x_i}/x_i^2}$ restricted to the relative interior of the polytope of feasible solutions. $P$trajectories of a canonical form linear program are radial projections of $A$trajectories of an associated standard form linear program. As a consequence there is a polynomial time linear programming algorithm using the affine scaling vector field of this associated linear program: This algorithm is essentially Karmarkar’s algorithm. These trajectories are studied in subsequent papers by two nonlinear changes of variables called Legendre transform coordinates and projective Legendre transform coordinates, respectively. It will be shown that $P$trajectories have an algebraic and a geometric interpretation. They are algebraic curves, and they are geodesics (actually distinguished chords) of a geometry isometric to a Hubert geometry on a polytope combinatorially dual to the polytope of feasible solutions. The $A$trajectories of strict standard form linear programs have similar interpretations: They are algebraic curves, and are geodesics of a geometry isometric to Euclidean geometry.

I. Adler, N. Karmarkar, G. Resende and S. Veiga, An implementation of Karmarkar’s algorithm for linear programming, preprint, Univ. of California, Berkeley, 1986.
 Kurt M. Anstreicher, A monotonic projective algorithm for fractional linear programming, Algorithmica 1 (1986), no. 4, 483–498. MR 880734, DOI https://doi.org/10.1007/BF01840458
 V. Arnold, Les méthodes mathématiques de la mécanique classique, Éditions Mir, Moscow, 1976 (French). Traduit du russe par Djilali Embarek. MR 0474391
 Earl R. Barnes, A variation on Karmarkar’s algorithm for solving linear programming problems, Math. Programming 36 (1986), no. 2, 174–182. MR 866987, DOI https://doi.org/10.1007/BF02592024
 D. A. Bayer and J. C. Lagarias, The nonlinear geometry of linear programming. I. Affine and projective scaling trajectories, Trans. Amer. Math. Soc. 314 (1989), no. 2, 499–526. MR 1005525, DOI https://doi.org/10.1090/S00029947198910055256
 H. Busemann and B. B. Phadke, A general version of Beltrami’s theorem in the large, Pacific J. Math. 115 (1984), no. 2, 299–315. MR 765189
 Herbert Busemann, The geometry of geodesics, Academic Press Inc., New York, N. Y., 1955. MR 0075623
 Herbert Busemann, Spaces with distinguished shortest joins, A spectrum of mathematics (Essays presented to H. G. Forder), Auckland Univ. Press, Auckland, 1971, pp. 108–120. MR 0431075 R. Courant and D. Hubert, Methods of mathematical physics, Vols. I, II, Wiley, New York, 1962.
 I. I. Dikin, Iterative solution of problems of linear and quadratic programming, Dokl. Akad. Nauk SSSR 174 (1967), 747–748 (Russian). MR 0221850 , About the convergence of an iterative process, Controllable Systems IM IK SO AN SSR 1974, No. 12, 5460. (Russian)
 W. Fenchel, On conjugate convex functions, Canad. J. Math. 1 (1949), 73–77. MR 28365, DOI https://doi.org/10.4153/cjm1949007x
 Anthony V. Fiacco and Garth P. McCormick, Nonlinear programming: Sequential unconstrained minimization techniques, John Wiley and Sons, Inc., New YorkLondonSydney, 1968. MR 0243831
 Wendell H. Fleming, Functions of several variables, AddisonWesley Publishing Co., Inc., Reading, MAss.London, 1965. MR 0174675 C. B. Garcia and W. I. Zangwill, Pathways to solutions, fixed points and equilibria, PrenticeHall, Englewood Cliffs, N. J., 1981.
 Clovis C. Gonzaga, An algorithm for solving linear programming problems in $O(n^3L)$ operations, Progress in mathematical programming (Pacific Grove, CA, 1987) Springer, New York, 1989, pp. 1–28. MR 982713 D. Hilbert, Grundlagen der Geometrie, 7th ed., Leipzig, 1930. (English transl., Foundations of geometry.) J. Hooker, The projective linear programming algorithm, Interfaces 16 (1986), 7590.
 Masao Iri and Hiroshi Imai, A multiplicative barrier function method for linear programming, Algorithmica 1 (1986), no. 4, 455–482. MR 880733, DOI https://doi.org/10.1007/BF01840457
 N. Karmarkar, A new polynomialtime algorithm for linear programming, Combinatorica 4 (1984), no. 4, 373–395. MR 779900, DOI https://doi.org/10.1007/BF02579150
 Narendra K. Karmarkar, Jeffrey C. Lagarias, Lev Slutsman, and Pyng Wang, Power series variants of Karmarkartype algorithms, AT&T Tech. J. 68 (1989), no. 3, 20–36. MR 1021069 S. Kapoor and P. M. Vaidya, Fast algorithms for convex quadratic programming and multicommodity flows, Proc. 18th ACM Sympos. on Theory of Computing, 1986, pp. 147159.
 Masakazu Kojima, Shinji Mizuno, and Akiko Yoshise, A primaldual interior point algorithm for linear programming, Progress in mathematical programming (Pacific Grove, CA, 1987) Springer, New York, 1989, pp. 29–47. MR 982714
 J. C. Lagarias, The nonlinear geometry of linear programming. III. Projective Legendre transform coordinates and Hilbert geometry, Trans. Amer. Math. Soc. 320 (1990), no. 1, 193–225. MR 1058199, DOI https://doi.org/10.1090/S00029947199010581990
 Cornelius Lanczos, The Variational Principles of Mechanics, Mathematical Expositions, no. 4, University of Toronto Press, Toronto, Ont., 1949. MR 0034139
 Nimrod Megiddo, On the complexity of linear programming, Advances in economic theory (Cambridge, MA, 1985) Econom. Soc. Monogr., vol. 12, Cambridge Univ. Press, Cambridge, 1989, pp. 225–268. MR 1117042
 Nimrod Megiddo, Pathways to the optimal set in linear programming, Progress in mathematical programming (Pacific Grove, CA, 1987) Springer, New York, 1989, pp. 131–158. MR 982720 N. Megiddo and M. Shub, Boundary behavior of interior point algorithms in linear programming, IBM Research Report RJ5319, Sept. 1986.
 J. L. Nazareth, Homotopy techniques in linear programming, Algorithmica 1 (1986), no. 4, 529–535. MR 880737, DOI https://doi.org/10.1007/BF01840461
 James Renegar, A polynomialtime algorithm, based on Newton’s method, for linear programming, Math. Programming 40 (1988), no. 1, (Ser. A), 59–93. MR 923697, DOI https://doi.org/10.1007/BF01580724
 R. T. Rockafellar, Conjugates and Legendre transforms of convex functions, Canadian J. Math. 19 (1967), 200–205. MR 213496, DOI https://doi.org/10.4153/CJM19670124 , Convex analysis, Princeton Univ. Press, Princeton, N. J., 1970.
 Michael Shub, On the asymptotic behavior of the projective rescaling algorithm for linear programming, J. Complexity 3 (1987), no. 3, 258–269. MR 919676, DOI https://doi.org/10.1016/0885064X%2887%2990015X
 Gy. Sonnevend, An “analytical centre” for polyhedrons and new classes of global algorithms for linear (smooth, convex) programming, System modelling and optimization (Budapest, 1985) Lect. Notes Control Inf. Sci., vol. 84, Springer, Berlin, 1986, pp. 866–875. MR 903521, DOI https://doi.org/10.1007/BFb0043914 , A new method for solving a set of linear (convex) inequalities and its application for identification and optimization, Proc. Sympos. on Dynamic Modelling, IFACIFORS, Budapest, June 1986.
 Josef Stoer and Christoph Witzgall, Convexity and optimization in finite dimensions. I, Die Grundlehren der mathematischen Wissenschaften, Band 163, SpringerVerlag, New YorkBerlin, 1970. MR 0286498 P. Vaidya, An algorithm for linear programming which requires $O(((m + n){n^2} + {(m + n)^{1.5}}n)L)$ arithmetic operations, Proc. 19th ACM Sympos. on Theory of Computing, 1987, pp. 2938.
 Robert J. Vanderbei, Marc S. Meketon, and Barry A. Freedman, A modification of Karmarkar’s linear programming algorithm, Algorithmica 1 (1986), no. 4, 395–407. MR 880730, DOI https://doi.org/10.1007/BF01840454
Retrieve articles in Transactions of the American Mathematical Society with MSC: 90C05, 34A99, 52A40
Retrieve articles in all journals with MSC: 90C05, 34A99, 52A40
Additional Information
Article copyright:
© Copyright 1989
American Mathematical Society