Abstract:
This paper examines the theoretical efficiency of solving a standard-form linear program by solving a sequence of shifted-barrier problems of the form minimize cTx - n (xj + ehj) j.,1 x s.t. Ax = b , x + e h > , for a given and fixed shift vector h > 0, and for a sequence of values of > 0 that converges to zero. The resulting sequence of solutions to the shifted barrier problems will converge to a solution to the standard form linear program. The advantage of using the shiftedbarrier approach is that a starting feasible solution is unnecessary, and there is no need for a Phase I-Phase II approach to solving the linear program, either directly or through the addition of an artificial variable. Furthermore, the algorithm can be initiated with a "warm start," i.e., an initial guess of a primal solution x that need not be feasible. The number of iterations needed to solve the linear program to a desired level of accuracy will depend on a measure of how close the initial solution x is to being feasible. The number of iterations will also depend on the judicious choice of the shift vector h . If an approximate center of the dual feasible region is known, then h can be chosen so that the guaranteed fractional decrease in e at each iteration is (1 - 1/(6 i)) , which contributes a factor of 6 ii to the number of iterations needed to solve the problem. The paper also analyzes the complexity of computing an approximate center of the dual feasible region from a "warm start," i.e., an initial (possibly infeasible) guess ir of a solution to the center problem of the dual. Key Words: linear program, interior-point algorithm, center, barrier function, shifted-barrier function, Newton step.