An Uncertainty-Aware Pseudo-Label Selection Framework using Regularized Conformal Prediction
This work addresses the challenge of domain-agnostic semi-supervised learning for researchers and practitioners, offering a method to enhance pseudo-labeling without relying on domain-specific augmentations, though it is incremental as it builds on existing conformal prediction techniques.
The paper tackles the problem of pseudo-labeling in semi-supervised learning, which underperforms due to poorly calibrated models, by proposing an uncertainty-aware framework using conformal regularization to reduce noisy training data, achieving improved performance in classification tasks.
Consistency regularization-based methods are prevalent in semi-supervised learning (SSL) algorithms due to their exceptional performance. However, they mainly depend on domain-specific data augmentations, which are not usable in domains where data augmentations are less practicable. On the other hand, Pseudo-labeling (PL) is a general and domain-agnostic SSL approach that, unlike consistency regularization-based methods, does not rely on the domain. PL underperforms due to the erroneous high-confidence predictions from poorly calibrated models. This paper proposes an uncertainty-aware pseudo-label selection framework that employs uncertainty sets yielded by the conformal regularization algorithm to fix the poor calibration neural networks, reducing noisy training data. The codes of this work are available at: https://github.com/matinmoezzi/ups conformal classification