BriefGPT.xyz
Oct, 2021
超网络用于连续的半监督学习
Hypernetworks for Continual Semi-Supervised Learning
HTML
PDF
Dhanajit Brahma, Vinay Kumar Verma, Piyush Rai
TL;DR
该研究提出了一种用于半监督连续学习的元一致性框架,该框架采用超网络来学习基础网络的半监督辅助分类器生成对抗网络的权重,并整合了超网络中连续任务的知识,以在新集合上进行测试。
Abstract
Learning from data sequentially arriving, possibly in a non i.i.d. way, with changing task distribution over time is called
continual learning
. Much of the work thus far in
continual learning
focuses on supervise
→