[Adapt] [Seminar] How to mitigate the requirement for labeled data? Self-Tuning for Data-Efficient Deep Learning.
Xiujie Song
songxj2018 at lzu.edu.cn
Tue Mar 22 19:33:35 CST 2022
Hi Adapters,
Deep learning has made revolutionary advances to diverse applications in the presence of large-scale labeled datasets. However, it is prohibitively time-costly and labor-expensive to collect sufficient labeled data in most realistic scenarios. How to mitigate the requirement for labeled data is one important problem.
To mitigate the requirement for labeled data, semi-supervised learning (SSL) focuses on simultaneously exploring both labeled and unlabeled data, while transfer learning (TL) popularizes a favorable practice of fine-tuning a pre-trained model to the target data. A dilemma is thus encountered: Without a decent pre-trained model to provide an implicit regularization, SSL through self-training from scratch will be easily misled by inaccurate pseudo-labels, especially in large-sized label space; Without exploring the intrinsic structure of unlabeled data, TL through fine-tuning from limited labeled data is at risk of under-transfer caused by model shift.
To escape from this dilemma, some researchers present Self-Tuning to enable data-efficient deep learning by unifying the exploration of labeled and unlabeled data and the transfer of a pre-trained model, as well as a Pseudo Group Contrast (PGC) mechanism to mitigate the reliance on pseudo-labels and boost the tolerance to false labels.
In this talk, I will introduce this paper, Self-Tuning for Data-Efficient Deep Learning, to you, which is published in ICML 2021. Hope you enjoy it~
For our online group seminar tomorrow, I prepared a Tencent meeting room:
###############################################################
边缘 邀请您参加腾讯会议
会议主题:边缘预定的会议
会议时间:2022/03/23 16:00-18:00 (GMT+08:00) 中国标准时间 - 北京
点击链接入会,或添加至会议列表:
https://meeting.tencent.com/dm/ASSrQvyjywtd
#腾讯会议:498-963-814
手机一键拨号入会
+8675536550000,,498963814# (中国大陆)
+85230018898,,,2,498963814# (中国香港)
根据您的位置拨号
+8675536550000 (中国大陆)
+85230018898 (中国香港)
复制该信息,打开手机腾讯会议即可参与
##############################################################
Time: Wed 4:00pm
Venue: Tencent meeting 498-963-814
Best Wishes,
Xiujie Song
More information about the Adapt
mailing list