Authors:
Kyung Ho Park
and
HyunHee Chung
Affiliation:
SOCAR, Republic of Korea
Keyword(s):
Label Ambiguity, Semi-Supervised Learning, Pseudo-Labeling, Uncertainty Estimation.
Abstract:
Upon the dominant accomplishments of deep neural networks, recent studies have scrutinized a robust model under the inherently ambiguous samples. Prior works have achieved superior performance under these ambiguous samples through label distribution approaches, assuming the existence of multiple human annotators. However, the aforementioned problem setting is not generally feasible due to resource constraints. For a generally applicable solution to the ambiguity problem, we propose Uncertainty-Guided Pseudo-Labeling (UGPL), a proof-of-concept level framework that leverages ambiguous samples on elevating the image recognition performance. Key contributions of our study are as follows. First, we constructed synthetic ambiguous datasets as there were no public benchmark dataset that deals with ambiguity problem. Given ambiguous samples, we empirically showed that not every ambiguous sample has meaningful knowledge consistent to the obvious samples at the target classes. We then examined
uncertainty can be a possible proxy for measuring the effectiveness of ambiguous sample’s knowledge toward the escalation of image recognition performance. Moreover, we validated pseudo-labeled ambiguous samples with low uncertainty better contributes to the test accuracy elevation. Lastly, we validated the UGPL showed larger accuracy elevation under the small size of obvious samples; thus, general practitioners can be widely benefited. To this end, we suggest further avenues of improvement practical techniques that resolve the ambiguity problem.
(More)