On the best choice of a damping sequence in iterative optimization methods

• Autores: Leonid Vaserstein
• Localización: Publicacions matematiques, ISSN 0214-1493, Vol. 32, Nº 2, 1988, págs. 275-287
• Idioma: inglés
• DOI: 10.5565/publmat_32288_11
• Títulos paralelos:
  • Sobre la mejor elección de una sucesión amortiguada en métodos de optimización iterativos
• Enlaces
  • Texto completo
• Resumen

  • Some iterative methods of mathematical programming use a damping sequence {at} such that 0 = at = 1 for all t, at ? 0 as t ? 8, and S at = 8. For example, at = 1/(t+1) in Brown's method for solving matrix games. In this paper, for a model class of iterative methods, the convergence rate for any damping sequence {at} depending only on time t is computed. The computation is used to find the best damping sequence.