Comparison of a special-purpose algorithm with general-purpose algorithms for solving geometric programming problems

We study the performance of four general-purpose nonlinear programming algorithms and one special-purpose geometric programming algorithm when used to solve geometric programming problems. Experiments are reported which show that the special-purpose algorithm GGP often finds approximate solutions more quickly than the general-purpose algorithm GRG2, but is usually not significantly more efficient than GRG2 when greater accuracy is required. However, for some of the most difficult test problems attempted, GGP was dramatically superior to all of the other algorithms. The other algorithms are usually not as efficient as GGP or GRG2. The ellipsoid algorithm is most robust.This work was supported in part by the National Science Foundation, Grant No. MCS-81-02141.     

Generalized permutation symmetry and the flavour problem inSU(2)L×U(1)

A generalized permutation group is introduced as a possible horizontal symmetry forSU(2)_L×U(1) gauge theories. It leads to the unique two generation quark mass matrices with a correct prediction for the Cabibbo angle. For three generations the model exhibits spontaneousCP violation, correlates the Kobayashi-Maskawa mixing parameterss ₁ ands ₃ and predicts an upper bound for the running top quark mass of approximately 45 GeV. The hierarchy of generations is due to a hierarchy of vacuum expectation values rather than of Yukawa coupling constants.     

Optimal multipliers for linear congruential pseudo-random number generators with prime moduli

A new method is described to find optimal multipliers for linear congruential pseudo-random number generation. The moduli especially considered are primes with primitive roots as multipliers. A table of such optimal multipliers is included.     

A reduced gradient method for quadratic programs with quadratic constraints and lp-constrained lp-approximation problems

A dual algorithm is developed for solving a general class of nonlinear programs that properly contains all convex quadratic programs with quadratic constraints and l_p-constrained l_p-approximation problems. The general dual program to be solved has essentially linear constraints but the objective function is nondifferentiable when certain variables equal zero. Modifications to the reduced gradient method for linearly constrained problems are presented that overcome the numerical difficulties associated with the nondifferentiable objective function. These modifications permit ‘blocks’ of variables to move to and away from zero on certain iterations even though the objective function is nondifferentiable at points having a block of variables equal to zero.     

Finding all efficient extreme points for multiple objective linear programs

In this paper we develop a method for finding all efficient extreme points for multiple objective linear programs. Simple characterizations of the efficiency of an edge incident to a nondegenerate or a degenerate efficient vertex are given. These characterizations form the basis of an algorithm for enumerating all efficient vertices. The algorithm appears to have definite computational advantages over other methods. Some illustrative examples are included.