Key points are not available for this paper at this time.
An algorithm for minimizing the largest eigenvalue of an affine combination of symmetric matrices is presented. The nonsmooth problem is transformed into an equivalent smooth constrained problem, which is solved by a predictor-corrector interior-point method taking full advantage of the differentiability and convexity. Some promising numerical results obtained from a preliminary implementation are included.
Florian Jarre (Wed,) studied this question.