Pseudo Labels for Single Positive Multi-Label Learning

Julio Arroyo

Keywords: Multi-Label Learning, Pseudo Labels, semi-supervised, image classification

TL;DR: To circumvent the difficulties of exhaustive annotation in multi-label image classification, we train a “teacher” model with minimal supervision, which then provides synthetic labels so that a “student” model learns on fully-labeled images.

Abstract: The cost of data annotation is a substantial impediment for multi-label image classification: in every image, every category must be labeled as present or absent. Single positive multi-label (SPML) learning is a cost-effective solution, where models are trained on a single positive label per image. Thus, SPML is a more challenging domain, since it requires dealing with missing labels. In this work, we propose a method to turn single positive data into fully-labeled data: “Pseudo Multi-Labels”. Basically, a “teacher” network is trained on single positive labels. Then, we treat the “teacher” model's predictions on the training data as ground-truth labels to train a “student” network on fully-labeled images. With this simple approach, we show that the performance achieved by the “student” model approaches that of a model trained on the actual fully-labeled images.