Seminar: Active Learning Meets Self-supervised Model: Effectiveness and Efficiency, 01st June, 1pm

When: Thursday 1st of June, 1pm AEST

Where: This seminar will be partially presented at the Rose Street Seminar area (J04) and partially online via Zoom. RSVP

Speaker: Ziting Wen

Title: Active Learning Meets Self-supervised Model: Effectiveness and Efficiency

Abstract:

The high cost of annotation plagues the application of deep learning models. Active learning has emerged as a classic approach to alleviate this issue by selecting the most informative samples to annotate. On the other hand, self-supervised training has proven to be effective in leveraging unlabeled data to generate powerful feature encoders. In this seminar, I will demonstrate how to integrate the strengths of both methodologies to design active learning strategies that operate on top of self-supervised models based on our recent work. By doing so, we aim to enhance the effectiveness and efficiency of the annotation process.

Bio:

Ziting Wen received his Master’s degree in information and communication engineering from Shanghai Jiao Tong University, China in 2020. Since then, he has been a PhD student in the marine group, ACFR. His research interests include active learning and semi-supervised learning.