Generalized Semi-Supervised Learning via Self-Supervised Feature Adaptation

TL;DR

This paper introduces Self-Supervised Feature Adaptation (SSFA), a framework that enhances semi-supervised learning by adapting features to mixed data distributions, leading to improved pseudo-label quality and performance across diverse scenarios.

Contribution

The paper proposes SSFA, a novel method that decouples pseudo-label prediction from the model and incorporates self-supervised tasks for feature adaptation in SSL.

Findings

SSFA improves SSL performance across various datasets.

SSFA enhances pseudo-label quality in mixed distribution scenarios.

SSFA outperforms existing SSL methods in experiments.

Abstract

Traditional semi-supervised learning (SSL) assumes that the feature distributions of labeled and unlabeled data are consistent which rarely holds in realistic scenarios. In this paper, we propose a novel SSL setting, where unlabeled samples are drawn from a mixed distribution that deviates from the feature distribution of labeled samples. Under this setting, previous SSL methods tend to predict wrong pseudo-labels with the model fitted on labeled data, resulting in noise accumulation. To tackle this issue, we propose Self-Supervised Feature Adaptation (SSFA), a generic framework for improving SSL performance when labeled and unlabeled data come from different distributions. SSFA decouples the prediction of pseudo-labels from the current model to improve the quality of pseudo-labels. Particularly, SSFA incorporates a self-supervised task into the SSL framework and uses it to adapt the feature extractor of the model to the unlabeled data. In this way, the extracted features better fit the distribution of unlabeled data, thereby generating high-quality pseudo-labels. Extensive experiments show that our proposed SSFA is applicable to various pseudo-label-based SSL learners and significantly improves performance in labeled, unlabeled, and even unseen distributions.