A Minimax Theorem with Applications to Machine Learning, Signal Processing, and Finance

Seung-Jean Kim(sjkimstanford.edu)
Stephen Boyd(boydstanford.edu)

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, 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: convex optimization, minimax theorem, portfolio optimization, robust beamforming, robust classification

Category 1: Convex and Nonsmooth Optimization (Convex Optimization )

Category 2: Applications -- OR and Management Sciences (Finance and Economics )

Category 3: Robust Optimization

Citation: Technical report, Stanford University, 10/07

Download: [PDF]

Entry Submitted: 11/13/2007
Entry Accepted: 11/13/2007
Entry Last Modified: 11/13/2007

Modify/Update this entry