Towards Resource-friendly, Extensible and Stable Incomplete Multi-view Clustering

Shengju Yu, Zhibin Dong, Siwei Wang, Xinhang Wan, Yue Liu, Weixuan Liang, Pei Zhang, Wenxuan Tu, Xinwang Liu
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:57415-57440, 2024.

Abstract

Incomplete multi-view clustering (IMVC) methods typically encounter three drawbacks: (1) intense time and/or space overheads; (2) intractable hyper-parameters; (3) non-zero variance results. With these concerns in mind, we give a simple yet effective IMVC scheme, termed as ToRES. Concretely, instead of self-expression affinity, we manage to construct prototype-sample affinity for incomplete data so as to decrease the memory requirements. To eliminate hyper-parameters, besides mining complementary features among views by view-wise prototypes, we also attempt to devise cross-view prototypes to capture consensus features for jointly forming high-quality clustering representation. To avoid the variance, we successfully unify representation learning and clustering operation, and directly optimize the discrete cluster indicators from incomplete data. Then, for the resulting objective function, we provide two equivalent solutions from perspectives of feasible region partitioning and objective transformation. Many results suggest that ToRES exhibits advantages against 20 SOTA algorithms, even in scenarios with a higher ratio of incomplete data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-yu24b, title = {Towards Resource-friendly, Extensible and Stable Incomplete Multi-view Clustering}, author = {Yu, Shengju and Dong, Zhibin and Wang, Siwei and Wan, Xinhang and Liu, Yue and Liang, Weixuan and Zhang, Pei and Tu, Wenxuan and Liu, Xinwang}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {57415--57440}, 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/yu24b/yu24b.pdf}, url = {https://proceedings.mlr.press/v235/yu24b.html}, abstract = {Incomplete multi-view clustering (IMVC) methods typically encounter three drawbacks: (1) intense time and/or space overheads; (2) intractable hyper-parameters; (3) non-zero variance results. With these concerns in mind, we give a simple yet effective IMVC scheme, termed as ToRES. Concretely, instead of self-expression affinity, we manage to construct prototype-sample affinity for incomplete data so as to decrease the memory requirements. To eliminate hyper-parameters, besides mining complementary features among views by view-wise prototypes, we also attempt to devise cross-view prototypes to capture consensus features for jointly forming high-quality clustering representation. To avoid the variance, we successfully unify representation learning and clustering operation, and directly optimize the discrete cluster indicators from incomplete data. Then, for the resulting objective function, we provide two equivalent solutions from perspectives of feasible region partitioning and objective transformation. Many results suggest that ToRES exhibits advantages against 20 SOTA algorithms, even in scenarios with a higher ratio of incomplete data.} }
Endnote
%0 Conference Paper %T Towards Resource-friendly, Extensible and Stable Incomplete Multi-view Clustering %A Shengju Yu %A Zhibin Dong %A Siwei Wang %A Xinhang Wan %A Yue Liu %A Weixuan Liang %A Pei Zhang %A Wenxuan Tu %A Xinwang Liu %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-yu24b %I PMLR %P 57415--57440 %U https://proceedings.mlr.press/v235/yu24b.html %V 235 %X Incomplete multi-view clustering (IMVC) methods typically encounter three drawbacks: (1) intense time and/or space overheads; (2) intractable hyper-parameters; (3) non-zero variance results. With these concerns in mind, we give a simple yet effective IMVC scheme, termed as ToRES. Concretely, instead of self-expression affinity, we manage to construct prototype-sample affinity for incomplete data so as to decrease the memory requirements. To eliminate hyper-parameters, besides mining complementary features among views by view-wise prototypes, we also attempt to devise cross-view prototypes to capture consensus features for jointly forming high-quality clustering representation. To avoid the variance, we successfully unify representation learning and clustering operation, and directly optimize the discrete cluster indicators from incomplete data. Then, for the resulting objective function, we provide two equivalent solutions from perspectives of feasible region partitioning and objective transformation. Many results suggest that ToRES exhibits advantages against 20 SOTA algorithms, even in scenarios with a higher ratio of incomplete data.
APA
Yu, S., Dong, Z., Wang, S., Wan, X., Liu, Y., Liang, W., Zhang, P., Tu, W. & Liu, X.. (2024). Towards Resource-friendly, Extensible and Stable Incomplete Multi-view Clustering. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:57415-57440 Available from https://proceedings.mlr.press/v235/yu24b.html.

Related Material