An Algorithm Based on Active Sets and Smoothing for Discretized Semi-Infinite Minimax Problems
- 195 Downloads
We present a new active-set strategy which can be used in conjunction with exponential (entropic) smoothing for solving large-scale minimax problems arising from the discretization of semi-infinite minimax problems. The main effect of the active-set strategy is to dramatically reduce the number of gradient calculations needed in the optimization. Discretization of multidimensional domains gives rise to minimax problems with thousands of component functions. We present an application to minimizing the sum of squares of the Lagrange polynomials to find good points for polynomial interpolation on the unit sphere in ℝ3. Our numerical results show that the active-set strategy results in a modified Armijo gradient or Gauss-Newton like methods requiring less than a quarter of the gradients, as compared to the use of these methods without our active-set strategy. Finally, we show how this strategy can be incorporated in an algorithm for solving semi-infinite minimax problems.
KeywordsMinimax problems Log-sum-exponential smoothing Active set strategies
Unable to display preview. Download preview PDF.
- 1.Outendijk, G.Z.: Methods of Feasible Directions. Elsevier, Amsterdam (1960) Google Scholar
- 2.Polak, E.: Computational Methods in Optimization: A Unified Approach. Academic, New York (1971) Google Scholar
- 3.Fletcher, R.: Practical Methods of Optimization. Wiley, New York (2000) Google Scholar
- 7.Kort, B.W., Bertsekas, D.P.: A new penalty function for constrained minimization. In: Proceedings 1972 IEEE Conference on Decision and Control, New Orleans, December 1972 Google Scholar
- 8.Bertsekas, D.P.: Nondifferentiable optimization via approximation. In: Balinski, M.L., Wolfe, P. (eds.) Mathematical Programming Study 3, Nondifferentiable Optimization, pp. 1–15. North-Holland, Amsterdam (1975) Google Scholar
- 10.Li, X.S.: An entropy-based aggregate method for minimax optimization. Eng. Optim. 14, 277–285 (1997) Google Scholar
- 14.Bertsekas, D.P.: Minmax methods based on approximation. In: Proceeding of the 1976 Johns Hopkins Conference on Information Sciences and Systems, Baltimore, 1976 Google Scholar
- 15.Bertsekas, D.P.: Lagrange Multiplier Methods in Constrained Optimization. Academic, New York (1982), Chap. 3; republished by Athena Scientific, Cambridge (1996) Google Scholar