发表机构
College of General Education, Chongqing Polytechnic University of Electronic Technology（重庆电子工程职业学院通识教育学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出单次调用Polyak外梯度方法SPolyakEG，通过回收外推评估降低计算成本，在单调和强单调条件下分别达到O(1/K)和线性收敛。
AI 中文摘要
我们提出了一种Polyak外梯度的单次调用变体，称为SPolyakEG，通过回收先前的外推算子评估来实现。在延迟临界条件下，我们证明了SPolyakEG对单调算子实现了$O(1/K)$残差收敛速率，并在强单调性下实现线性收敛。与PolyakEG相比，SPolyakEG将每次迭代的算子评估次数从两次减少到一次。
英文摘要
We propose a single-call variant of Polyak extragradient, termed SPolyakEG, by recycling the previous extrapolation operator evaluation. Under a delayed critical condition, we establish that SPolyakEG achieves an $O(1/K)$ residual convergence rate for monotone operators and linear convergence under strong monotonicity. Compared with PolyakEG, SPolyakEG reduces the number of operator evaluations from two to one per iteration.
