Code of "HSACC: Incomplete Multi-view Clustering via Hierarchical Semantic Alignment and Cooperative Completion" (NeurIPS 2025)

torch==1.12.0

tensorflow==2.10.0

numpy>=1.21.0

scikit-learn>=0.22.0

munkres>=1.1.4

In the code, we have provided hyperparameter settings for the Caltech101-20, NoisyMNIST, and LandUse_21 datasets. Below, we additionally include the hyperparameter configurations for the Hdigit and 100leaves datasets.

Hdigit

lambda1 = 0.1 lambda2 = 0.1 lambda3 = 10 lambda4 = 1

100leaves

lambda1 = 0.1 lambda2 = 0.1 lambda3 = 10 lambda4 = 1

The code includes example implementations for both two-view and multi-view settings, which can be run via run.py (two-view) and run_multiview.py (multi-view).

You can adjust the missing rate for the dataset by using the --missing_rate argument. For example, --missing_rate 0.5 sets 50% of the data as missing. This allows you to run experiments under different levels of missing data.

If you find HSACC useful in your research, please consider giving us a star and citing it with the following BibTeX entry:

@misc{ding2025incompletemultiviewclusteringhierarchical,
  title={Incomplete Multi-view Clustering via Hierarchical Semantic Alignment and Cooperative Completion}, 
  author={Xiaojian Ding and Lin Zhao and Xian Li and Xiaoying Zhu},
  year={2025},
  eprint={2510.13887},
  archivePrefix={arXiv},
  primaryClass={eess.IV},
  url={https://arxiv.org/abs/2510.13887}
}