BriefGPT.xyz
May, 2023
在在线连续学习中处理跨任务类别歧视
Dealing with Cross-Task Class Discrimination in Online Continual Learning
HTML
PDF
Yiduo Guo, Bing Liu, Dongyan Zhao
TL;DR
本文提出了一个新的优化目标和基于梯度调整方法的动态训练偏差方法,以解决回放方法在在线连续学习过程中的动态训练偏差问题,并在实验中取得了更好的结果,以解决跨任务类别辨别问题。
Abstract
Existing
continual learning
(CL) research regards catastrophic forgetting (CF) as almost the only challenge. This paper argues for another challenge in
class-incremental learning
(CIL), which we call
→