Learning with Partial-Label and Unlabeled Data: A Uniform Treatment for Supervision Redundancy and Insufficiency

Yangfan Liu, Jiaqi Lv, Xin Geng, Ning Xu
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:31614-31628, 2024.

Abstract

One major challenge in weakly supervised learning is learning from inexact supervision, ranging from partial labels (PLs) with redundant information to the extreme of unlabeled data with insufficient information. While recent work has made significant strides in specific inexact supervision contexts, supervision forms typically coexist in complex combinations. This is exemplified in semi-supervised partial label learning, where PLs act as the exclusive supervision in a semi-supervised setting. Current strategies addressing combined inexact scenarios are usually composite, which can lead to incremental solutions that essentially replicate existing methods. In this paper, we propose a novel approach to uniformly tackle both label redundancy and insufficiency, derived from a mutual information-based perspective. We design a label channel that facilitates dynamic label exchange within the candidate label sets, which identifies potential true labels and filters out likely incorrect ones, thereby minimizing error accumulation. Experimental results demonstrate the superiority of our method over existing state-of-the-art PL and semi-supervised learning approaches by directly integrating them. Furthermore, our extended experiments on partial-complementary label learning underscore the flexibility of our uniform treatment in managing diverse supervision scenarios.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-liu24ar, title = {Learning with Partial-Label and Unlabeled Data: A Uniform Treatment for Supervision Redundancy and Insufficiency}, author = {Liu, Yangfan and Lv, Jiaqi and Geng, Xin and Xu, Ning}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {31614--31628}, 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/liu24ar/liu24ar.pdf}, url = {https://proceedings.mlr.press/v235/liu24ar.html}, abstract = {One major challenge in weakly supervised learning is learning from inexact supervision, ranging from partial labels (PLs) with redundant information to the extreme of unlabeled data with insufficient information. While recent work has made significant strides in specific inexact supervision contexts, supervision forms typically coexist in complex combinations. This is exemplified in semi-supervised partial label learning, where PLs act as the exclusive supervision in a semi-supervised setting. Current strategies addressing combined inexact scenarios are usually composite, which can lead to incremental solutions that essentially replicate existing methods. In this paper, we propose a novel approach to uniformly tackle both label redundancy and insufficiency, derived from a mutual information-based perspective. We design a label channel that facilitates dynamic label exchange within the candidate label sets, which identifies potential true labels and filters out likely incorrect ones, thereby minimizing error accumulation. Experimental results demonstrate the superiority of our method over existing state-of-the-art PL and semi-supervised learning approaches by directly integrating them. Furthermore, our extended experiments on partial-complementary label learning underscore the flexibility of our uniform treatment in managing diverse supervision scenarios.} }
Endnote
%0 Conference Paper %T Learning with Partial-Label and Unlabeled Data: A Uniform Treatment for Supervision Redundancy and Insufficiency %A Yangfan Liu %A Jiaqi Lv %A Xin Geng %A Ning Xu %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-liu24ar %I PMLR %P 31614--31628 %U https://proceedings.mlr.press/v235/liu24ar.html %V 235 %X One major challenge in weakly supervised learning is learning from inexact supervision, ranging from partial labels (PLs) with redundant information to the extreme of unlabeled data with insufficient information. While recent work has made significant strides in specific inexact supervision contexts, supervision forms typically coexist in complex combinations. This is exemplified in semi-supervised partial label learning, where PLs act as the exclusive supervision in a semi-supervised setting. Current strategies addressing combined inexact scenarios are usually composite, which can lead to incremental solutions that essentially replicate existing methods. In this paper, we propose a novel approach to uniformly tackle both label redundancy and insufficiency, derived from a mutual information-based perspective. We design a label channel that facilitates dynamic label exchange within the candidate label sets, which identifies potential true labels and filters out likely incorrect ones, thereby minimizing error accumulation. Experimental results demonstrate the superiority of our method over existing state-of-the-art PL and semi-supervised learning approaches by directly integrating them. Furthermore, our extended experiments on partial-complementary label learning underscore the flexibility of our uniform treatment in managing diverse supervision scenarios.
APA
Liu, Y., Lv, J., Geng, X. & Xu, N.. (2024). Learning with Partial-Label and Unlabeled Data: A Uniform Treatment for Supervision Redundancy and Insufficiency. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:31614-31628 Available from https://proceedings.mlr.press/v235/liu24ar.html.

Related Material