抄録
This paper is concerned with improvement of the convergence rate of the computational algorithm using series expansion for the inverse operator of a positive-definite operator.
The fundamental operator polynomial Pm(A) is introduced, which is based on the first m-terms of the series expansion of the inverse of the Hessian of a quadratic performance index. By computing the operator polynomial Pm(Ai) at each iteration, a new computational algorithm is developed and the ratio of convergence for this algorithm is discussed.
This series expansion technique can also be applied to some optimal problem with constraints.
