A Minimax Theorem with Applications to Machine Learning, Signal Processing, and Finance
Authors:
- Seung-Jean KimSKSeung-Jean Kim
- Stephen BoydSBStephen Boyd
- 1 January 2008
- journal article
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Optimization
- Vol. 19 (3) , 1344-1367
Abstract
This paper concerns a fractional function of the form $x^Ta/\sqrt{x^TBx}$, where B is positive definite. We consider the game of choosing x from a convex set, to maximize the function, and choosing $(a,B)$ from a convex set, to minimize it. We prove the existence of a saddle point and describe an efficient method, based on convex optimization, for computing it. We describe applications in machine learning (robust Fisher linear discriminant analysis), signal processing (robust beamforming and robust matched filtering), and finance (robust portfolio selection). In these applications, x corresponds to some design variables to be chosen, and the pair $(a,B)$ corresponds to the statistical model, which is uncertain.
Keywords
This publication has 21 references indexed in Scilit:
- Optimal Inequalities in Probability Theory: A Convex Optimization ApproachSIAM Journal on Optimization, 2005
- Worst-Case Value-At-Risk and Robust Portfolio Optimization: A Conic Programming ApproachOperations Research, 2003
- An Interior-Point Method for a Class of Saddle-Point ProblemsJournal of Optimization Theory and Applications, 2003
- Robust Portfolio Selection ProblemsMathematics of Operations Research, 2003
- Robust portfolio selection using linear-matrix inequalitiesJournal of Economic Dynamics and Control, 2002
- Global Portfolio OptimizationCFA Magazine, 1992
- On the Sensitivity of Mean-Variance-Efficient Portfolios to Changes in Asset Means: Some Analytical and Computational ResultsThe Review of Financial Studies, 1991
- The Markowitz Optimization Enigma: Is ‘Optimized’ Optimal?CFA Magazine, 1989
- Bayes-Stein Estimation for Portfolio AnalysisJournal of Financial and Quantitative Analysis, 1986
- Robust techniques for signal processing: A surveyProceedings of the IEEE, 1985