Technion – Israel Institute of Technology
Title: Robust Solutions of Uncertain Optimization Problems Under Ambiguous Stochastic Data
Author(s): Aharon Ben-Tal
Abstract: We show how robust optimization (RO) can provide tractable, safe approximation to probabilistic constraints (chance constraints) even under partial information (ambiguity) on the random parameters.
In particular, we address the case where the only available information is on means and dispersion measures. Unlike previous attempts where the dispersion measure is the variance, here we derive tight approximations when the dispersion measure is the MAD (mean absolute deviation). The theory is applied to problems in portfolio selection, inventory management and antenna array design.