Open-Domain Semi-Supervised Learning via Glocal Cluster Structure Exploitation
- 22 February 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 36 (8) , 4163-4177
Abstract
Semi-supervised learning (SSL) aims to reduce the heavy reliance of current deep models on costly manual annotation by leveraging a large amount of unlabeled data in combination with a much smaller set of labeled data. However, most existing SSL methods assume that all labeled and unlabeled data are drawn from the same feature distribution, which can be impractical in real-world applications. In this study, we take the initial step to systematically investigate the open-domain semi-supervised learning setting, where a feature distribution mismatch exists between labeled and unlabeled data. In pursuit of an effective solution for open-domain SSL, we propose a novel framework called GlocalMatch , which aims to exploit both glo bal and lo cal ( i.e. , glocal) cluster structure of open-domain unlabeled data. The glocal cluster structure is utilized in two complementary ways. Firstly, GlocalMatch optimizes a Glocal Cluster Compacting (GCC) objective, that encourages feature representations of the same class, whether with in the same domain or across different domains, to become closer to each other. Secondly, GlocalMatch incorporates a Glocal Semantic Aggregation (GSA) strategy to produce more reliable pseudo-labels by aggregating predictions from neighboring clusters. Extensive experiments demonstrate that GlocalMatch outperforms the state-of-the-art SSL methods significantly, achieving superior performance for both in-domain and out-of-domain generalization. The code is released in https://github.com/nukezil/GlocalMatch .Keywords
Funding Information
- National Natural Science Foundation of China (62222604, 62206052, 62192783)
- Natural Science Foundation of Jiangsu Province (BK20210224)
This publication has 45 references indexed in Scilit:
- Towards Realistic Semi-supervised Learning2022
- Class-Aware Contrastive Semi-Supervised Learning2022
- Prototypical Cross-domain Self-supervised Learning for Few-shot Unsupervised Domain Adaptation2021
- Cross-Domain Gradient Discrepancy Minimization for Unsupervised Domain Adaptation2021
- Online Deep Clustering for Unsupervised Representation Learning2020
- A survey on semi-supervised learningMachine Learning, 2019
- Contrastive Adaptation Network for Unsupervised Domain Adaptation2019
- Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation2018
- Unsupervised Feature Learning via Non-parametric Instance Discrimination2018
- Adversarial Discriminative Domain Adaptation2017