MGO-SC：面向噪声深度学习的动量引导优化器一致性门控
原标题：Moment-Guided Optimizers for Noisy Deep Learning: MGO-SC as a Simple Consistency Gate for Adam-Style Updates
AI 摘要
该论文研究MGO-SC，一种在Adam类优化器中加入符号一致性门控的轻量改进方法，用于应对标签噪声下梯度符号不稳定的问题。作者发现朴素的收缩式门控会损失有效步长并劣于AdamW基线，因此提出保持均值的变体，在不改变每张量净步长幅度的前提下重新分配坐标信任度。在CIFAR-10与ResNet-18上，该方法在干净、对称和非对称标签噪声下均稳定优于AdamW（如40%对称噪声下提升1.4个百分点，配对bootstrap置信区间不含零，Wilcoxon p<0.01），且不增加二阶信息或逐样本过滤，代码与原始日志已开源。
正文节选
Moment-Guided Optimizers for Noisy Deep Learning: MGO-SC as a Simple Consistency Gate for Adam-Style Updates Abstract Adaptive optimizers are widely used in deep learning, yet their behavior can degrade under heavy stochasticity and label noise where the sign and magnitude of gradients can be unstable. We study MGO-SC (Moment-Guided Optimizer with Sign-Consistency), a lightweight modification of Adam-style methods that derives a coordinate-wise consistency gate from the magnitude of a sign expon