NP-Match: Towards a New Probabilistic Model for Semi-Supervised Learning

TL;DR

NP-Match introduces a novel semi-supervised learning method that adapts neural processes to improve pseudo-label quality and uncertainty estimation with lower computational costs, outperforming existing approaches.

Contribution

This work is the first to adapt neural processes for semi-supervised image classification, enhancing pseudo-label accuracy and uncertainty estimation efficiency.

Findings

Outperforms state-of-the-art SSL methods on multiple datasets.

Reduces computational overhead compared to MC dropout.

Effective across standard, imbalanced, and multi-label SSL tasks.

Abstract

Semi-supervised learning (SSL) has been widely explored in recent years, and it is an effective way of leveraging unlabeled data to reduce the reliance on labeled data. In this work, we adjust neural processes (NPs) to the semi-supervised image classification task, resulting in a new method named NP-Match. NP-Match is suited to this task for two reasons. Firstly, NP-Match implicitly compares data points when making predictions, and as a result, the prediction of each unlabeled data point is affected by the labeled data points that are similar to it, which improves the quality of pseudo-labels. Secondly, NP-Match is able to estimate uncertainty that can be used as a tool for selecting unlabeled samples with reliable pseudo-labels. Compared with uncertainty-based SSL methods implemented with Monte-Carlo (MC) dropout, NP-Match estimates uncertainty with much less computational overhead, which can save time at both the training and the testing phases. We conducted extensive experiments on five public datasets under three semi-supervised image classification settings, namely, the standard semi-supervised image classification, the imbalanced semi-supervised image classification, and the multi-label semi-supervised image classification, and NP-Match outperforms state-of-the-art (SOTA) approaches or achieves competitive results on them, which shows the effectiveness of NP-Match and its potential for SSL. The codes are at https://github.com/Jianf-Wang/NP-Match