Generalized AFBA algorithm for saddle-point problems

The convex-concave minimax problem, known as the saddle-point problem, has been extensively studied from various aspects including the algorithm design, convergence condition and complexity. In this paper, we propose a generalized asymmetric forward-backward-adjoint (G-AFBA) algorithm to solve such a problem by utilizing both the proximal techniques and the interactive information of primal-dual updates. Except enjoying … Read more