Inexact smoothing method for large scale minimax optimization (Q425468)

The authors consider the finite minimax optimization problem \[ \min_{x\in\mathbb{R}^n}\,F(x),\quad\text{where }F(x)= \max_{j\in 1_m}\,f^j(x). \] For this problem, the authors propose a smoothing method based on a maximum entropy function and an inexact Newton-type algorithm for its solution. The algorithm is shown to both globally and superlinearly convergent. Numerical results are presented.