Realistic Evaluation of Deep Semi-Supervised Learning Algorithms

TL;DR

This paper critically evaluates deep semi-supervised learning algorithms in realistic scenarios, revealing limitations of current benchmarks and emphasizing the importance of robustness and proper baseline reporting.

Contribution

It provides a unified reimplementation and evaluation platform for SSL methods, highlighting issues like out-of-class data sensitivity and underreported baseline performances.

Findings

Simple baselines are often underreported in performance.

SSL methods vary in sensitivity to labeled and unlabeled data.

Performance drops significantly with out-of-class unlabeled data.

Abstract

Semi-supervised learning (SSL) provides a powerful framework for leveraging unlabeled data when labels are limited or expensive to obtain. SSL algorithms based on deep neural networks have recently proven successful on standard benchmark tasks. However, we argue that these benchmarks fail to address many issues that these algorithms would face in real-world applications. After creating a unified reimplementation of various widely-used SSL techniques, we test them in a suite of experiments designed to address these issues. We find that the performance of simple baselines which do not use unlabeled data is often underreported, that SSL methods differ in sensitivity to the amount of labeled and unlabeled data, and that performance can degrade substantially when the unlabeled dataset contains out-of-class examples. To help guide SSL research towards real-world applicability, we make our unified reimplemention and evaluation platform publicly available.