A Square Peg in a Square Hole: Meta-Expert for Long-Tailed Semi-Supervised Learning

TL;DR

This paper introduces a dynamic expert assignment approach for long-tailed semi-supervised learning, effectively utilizing multiple experts and multi-depth features to improve pseudo-label quality and reduce bias, especially under distribution mismatch.

Contribution

It proposes a novel dynamic expert assignment module and multi-depth feature fusion to better handle long-tailed semi-supervised learning with distribution mismatch.

Findings

Improved pseudo-label quality through expert specialization.

Theoretical analysis shows reduced generalization error.

Demonstrated effectiveness on multiple long-tailed datasets.

Abstract

This paper studies the long-tailed semi-supervised learning (LTSSL) with distribution mismatch, where the class distribution of the labeled training data follows a long-tailed distribution and mismatches with that of the unlabeled training data. Most existing methods introduce auxiliary classifiers (experts) to model various unlabeled data distributions and produce pseudo-labels, but the expertises of various experts are not fully utilized. We observe that different experts are good at predicting different intervals of samples, e.g., long-tailed expert is skilled in samples located in the head interval and uniform expert excels in samples located in the medium interval. Therefore, we propose a dynamic expert assignment module that can estimate the class membership (i.e., head, medium, or tail class) of samples, and dynamically assigns suitable expert to each sample based on the estimated membership to produce high-quality pseudo-label in the training phase and produce prediction in the testing phase. We also theoretically reveal that integrating different experts' strengths will lead to a smaller generalization error bound. Moreover, we find that the deeper features are more biased toward the head class but with more discriminative ability, while the shallower features are less biased but also with less discriminative ability. We, therefore, propose a multi-depth feature fusion module to utilize different depth features to mitigate the model bias. Our method demonstrates its effectiveness through comprehensive experiments on the CIFAR-10-LT, STL-10-LT, and SVHN-LT datasets across various settings.