Peng, Bohua;
Islam, Mobarakol;
Tu, Mei;
(2022)
Angular Gap: Reducing the Uncertainty of Image Difficulty through Model Calibration.
In:
Proceedings of the 30th ACM International Conference on Multimedia.
(pp. pp. 979-987).
ACM
Preview |
Text
2207.08525.pdf - Accepted Version Download (2MB) | Preview |
Abstract
Curriculum learning needs example difficulty to proceed from easy to hard. However, the credibility of image difficulty is rarely investigated, which can seriously affect the effectiveness of curricula. In this work, we propose Angular Gap, a measure of difficulty based on the difference in angular distance between feature embeddings and class-weight embeddings built by hyperspherical learning. To ascertain difficulty estimation, we introduce class-wise model calibration, as a post-training technique, to the learnt hyperbolic space. This bridges the gap between probabilistic model calibration and angular distance estimation of hyperspherical learning. We show the superiority of our calibrated Angular Gap over recent difficulty metrics on CIFAR10-H and ImageNetV2. We further propose a curriculum based on Angular Gap for unsupervised domain adaptation that can translate from learning easy samples to mining hard samples. We combine this curriculum with a state-of-the-art self-training method, Cycle Self Training (CST). The proposed Curricular CST learns robust representations and outperforms recent baselines on Office31 and VisDA 2017.
| Type: | Proceedings paper |
|---|---|
| Title: | Angular Gap: Reducing the Uncertainty of Image Difficulty through Model Calibration |
| Event: | MM '22: The 30th ACM International Conference on Multimedia |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1145/3503161.3548289 |
| Publisher version: | https://doi.org/10.1145/3503161.3548289 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Med Phys and Biomedical Eng |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10161581 |
Loading...
Loading...Loading...Loading...Archive Staff Only
 |
View Item |
