Official Implementation of the paper: Learning Clustering-Friendly Representations via Partial Information Discrimination and Cross-Level Interaction.
Congratulations!!! Our paper was officially accepted by the journal NEURAL NETWORKS (CCF B)
An overview of our PICI framework, which encompasses three partial information learning modules, namely, (a) the PISD module, which enforces the partial information self-discrimination upon the masked images via a Transformer auto-encoder, (b) the PICD module, which takes the class tokens [CLS] as input and performs two levels of contrastive learning, and (c) the CLI module, which enables the mutual interaction between the instance- and cluster-level subspaces by constraining their cross-level consistency.
There is a configuration file "config/config.yaml", which can be edited for both the training and test options.
After setting the configuration, to start training, simply run
python train_pici.pyAfter the training phase, there will be a saved model in the "model_path" specified in the configuration file. To validate the trained model, run
python cluster_pici.py@article{zhang2024learning,
title={Learning Representations for Clustering via Partial Information Discrimination and Cross-Level Interaction},
author={Zhang, Hai-Xin and Huang, Dong and Ling, Hua-Bao and Zhang, Guang-Yu and Sun, Wei-jun and Wen, Zi-hao},
journal={arXiv preprint arXiv:2401.13503},
year={2024}
}