论文标题
针对马鞍点问题的原始偶对序列子空间优化
Primal-Dual Sequential Subspace Optimization for Saddle-point Problems
论文作者
论文摘要
我们引入了一种新的顺序子空间优化方法,以解决大规模的鞍点问题。它在低维子空间中迭代地求解了辅助鞍点问题的序列,该序列由从原始\ emph {and}双变量的一阶信息得出的方向跨越。进一步部署近端正规化以稳定优化过程。实验结果表明,相对于流行的一阶方法,收敛明显更好。我们分析了子空间对算法融合的影响,并在各种确定性优化方案中评估其性能,例如Bi-linear Games,基于ADMM的约束优化和生成的对抗网络。
We introduce a new sequential subspace optimization method for large-scale saddle-point problems. It solves iteratively a sequence of auxiliary saddle-point problems in low-dimensional subspaces, spanned by directions derived from first-order information over the primal \emph{and} dual variables. Proximal regularization is further deployed to stabilize the optimization process. Experimental results demonstrate significantly better convergence relative to popular first-order methods. We analyze the influence of the subspace on the convergence of the algorithm, and assess its performance in various deterministic optimization scenarios, such as bi-linear games, ADMM-based constrained optimization and generative adversarial networks.