Li, Xing-Si; Fang, Shu-Cherng - In: Mathematical Methods of Operations Research 46 (1997) 1, pp. 119-130
Consider a min-max problem in the form of min<Subscript> xεX </Subscript>max<Subscript>1≤i≤m </Subscript>{f <Subscript> i </Subscript>(x)}. It is well-known that the non-differentiability of the max functionF(x) ≡ max<Subscript>1≤i≤m </Subscript>{f <Subscript> i </Subscript>(x)} presents difficulty in finding an optimal solution. An entropic regularization procedure provides a smooth...</subscript></subscript></subscript></subscript></subscript>