In this paper,we propose a mirror descent gradient ascent algorithm to solve one side relatively smooth nonconvex-concave minimax optimization problems.At each iteration of the proposed algorithm,a mirror descent step is performed to update the relatively smooth variable,while a gradient ascent projection step is used to update the smooth variable alternately.We also prove that the iteration complexity of the proposed algorithm is O(ε-4)to achieve an e-approximate first-order stationary point.
维普
万方数据