BriefGPT.xyz
Sep, 2023
小数据集有效训练的替代正则化方法
No Data Augmentation? Alternative Regularizations for Effective Training on Small Datasets
HTML
PDF
Lorenzo Brigato, Stavroula Mougiakakou
TL;DR
通过使用替代的正则化策略来推动小图像分类数据集上的有监督学习，研究取得了与最先进的方法相当的测试准确率。
Abstract
Solving
image classification
tasks given
small training datasets
remains an open challenge for modern computer vision. Aggressive
data augmentati
→