论文信息 - A Modified Conjugate Gradient Algorithm for Unconstrained Nonlinear Optimization

A Modified Conjugate Gradient Algorithm for Unconstrained Nonlinear Optimization

W E PRESENT a modification of the Fletcher-Reeves (FRCG) conjugate gradient algorithm [3]. This modification results in a more stable computational performance than the ones exhibited by either FRCG or PRCG (Polak-Ribiere conjugate gradient algorithm [7, 10]). We attribute this stable behavior to the modification proposed here, which incorporates one of the most important features of the quasi-Newton algorithm into the conjugate gradient procedure (this feature is missing in FRCG and PRCG) without increasing storage requirements. We shall present the motivation for the modification and support these arguments by a sample of encouraging computational results.

A. Perry

[1] H. H. Rosenbrock,et al. An Automatic Method for Finding the Greatest or Least Value of a Function , 1960, Comput. J..

[2] C. M. Reeves,et al. Function minimization by conjugate gradients , 1964, Comput. J..

[3] H. Y. Huang. Unified approach to quadratically convergent algorithms for function minimization , 1970 .

[4] R. Fletcher,et al. A New Approach to Variable Metric Algorithms , 1970, Comput. J..

[5] D. Shanno. Conditioning of Quasi-Newton Methods for Function Minimization , 1970 .

[6] D. Goldfarb. A family of variable-metric methods derived by variational means , 1970 .

[7] M. J. D. Powell,et al. Recent advances in unconstrained optimization , 1971, Math. Program..

[8] C. D. Mote,et al. Optimization methods for engineering design , 1971 .

[9] E. Polak,et al. Efficient Implementations of the Polak–Ribière Conjugate Gradient Algorithm , 1972 .

[10] M. J. D. Powell,et al. Restart procedures for the conjugate gradient method , 1977, Math. Program..