Correlation-Induced Label Prior for Semi-Supervised Multi-Label Learning

Biao Liu, Ning Xu, Xiangyu Fang, Xin Geng
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:32224-32238, 2024.

Abstract

Semi-supervised multi-label learning (SSMLL) aims to address the challenge of limited labeled data availability in multi-label learning (MLL) by leveraging unlabeled data to improve the model’s performance. Due to the difficulty of estimating the reliable label correlation on minimal multi-labeled data, previous SSMLL methods fail to unlash the power of the correlation among multiple labels to improve the performance of the predictive model in SSMLL. To deal with this problem, we propose a novel SSMLL method named PCLP where the correlation-induced label prior is inferred to enhance the pseudo-labeling instead of dirtily estimating the correlation among labels. Specifically, we construct the correlated label prior probability distribution using structural causal model (SCM), constraining the correlations of generated pseudo-labels to conform to the prior, which can be integrated into a variational label enhancement framework optimized by both labeled and unlabeled instances in a unified manner. Theoretically, we demonstrate the accuracy of the generated pseudo-labels and guarantee the learning consistency of the proposed method. Comprehensive experiments on several benchmark datasets have validated the superiority of the proposed method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-liu24bt, title = {Correlation-Induced Label Prior for Semi-Supervised Multi-Label Learning}, author = {Liu, Biao and Xu, Ning and Fang, Xiangyu and Geng, Xin}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {32224--32238}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/liu24bt/liu24bt.pdf}, url = {https://proceedings.mlr.press/v235/liu24bt.html}, abstract = {Semi-supervised multi-label learning (SSMLL) aims to address the challenge of limited labeled data availability in multi-label learning (MLL) by leveraging unlabeled data to improve the model’s performance. Due to the difficulty of estimating the reliable label correlation on minimal multi-labeled data, previous SSMLL methods fail to unlash the power of the correlation among multiple labels to improve the performance of the predictive model in SSMLL. To deal with this problem, we propose a novel SSMLL method named PCLP where the correlation-induced label prior is inferred to enhance the pseudo-labeling instead of dirtily estimating the correlation among labels. Specifically, we construct the correlated label prior probability distribution using structural causal model (SCM), constraining the correlations of generated pseudo-labels to conform to the prior, which can be integrated into a variational label enhancement framework optimized by both labeled and unlabeled instances in a unified manner. Theoretically, we demonstrate the accuracy of the generated pseudo-labels and guarantee the learning consistency of the proposed method. Comprehensive experiments on several benchmark datasets have validated the superiority of the proposed method.} }
Endnote
%0 Conference Paper %T Correlation-Induced Label Prior for Semi-Supervised Multi-Label Learning %A Biao Liu %A Ning Xu %A Xiangyu Fang %A Xin Geng %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-liu24bt %I PMLR %P 32224--32238 %U https://proceedings.mlr.press/v235/liu24bt.html %V 235 %X Semi-supervised multi-label learning (SSMLL) aims to address the challenge of limited labeled data availability in multi-label learning (MLL) by leveraging unlabeled data to improve the model’s performance. Due to the difficulty of estimating the reliable label correlation on minimal multi-labeled data, previous SSMLL methods fail to unlash the power of the correlation among multiple labels to improve the performance of the predictive model in SSMLL. To deal with this problem, we propose a novel SSMLL method named PCLP where the correlation-induced label prior is inferred to enhance the pseudo-labeling instead of dirtily estimating the correlation among labels. Specifically, we construct the correlated label prior probability distribution using structural causal model (SCM), constraining the correlations of generated pseudo-labels to conform to the prior, which can be integrated into a variational label enhancement framework optimized by both labeled and unlabeled instances in a unified manner. Theoretically, we demonstrate the accuracy of the generated pseudo-labels and guarantee the learning consistency of the proposed method. Comprehensive experiments on several benchmark datasets have validated the superiority of the proposed method.
APA
Liu, B., Xu, N., Fang, X. & Geng, X.. (2024). Correlation-Induced Label Prior for Semi-Supervised Multi-Label Learning. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:32224-32238 Available from https://proceedings.mlr.press/v235/liu24bt.html.

Related Material