Robust Graph Contrastive Learning for Incomplete Multi-view Clustering

Authors: Deyin Zhuang, Jian Dai, Xingfeng Li, Xi Wu, Yuan Sun, Zhenwen Ren

IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on the six multi-view datasets demonstrate that our RGCL exhibits superiority and effectiveness compared with 9 state-of-the-art IMVC methods. The source code is available at https://github.com/DYZ163/RGCL.git.
Researcher Affiliation Academia Deyin Zhuang1 , Jian Dai2 , Xingfeng Li1 , Xi Wu1 , Yuan Sun3,4 , Zhenwen Ren1 1Southwest University of Science and Technology, China 2Southwest Automation Research Institute, China 3 College of Computer Science, Sichuan University, China 4 National Key Laboratory of Fundamental Algorithms and Models for Engineering Numerical Simulation, Sichuan University, China EMAIL, EMAIL
Pseudocode No The paper describes the methodology in detail in Section 3, including Multi-view Reconstruction, Noise-robust Graph Contrastive Learning, Cross-view Graph-level Alignment, and Implementation, but it does not present any structured pseudocode or algorithm blocks.
Open Source Code Yes The source code is available at https://github.com/DYZ163/RGCL.git.
Open Datasets Yes In this section, we evaluate the performance of the proposed method on the six multi-view datasets, including Hand Written [Le Cun et al., 1989], COIL20 [Nene et al., 1996], BDGP [Cai et al., 2012], Land Use-21 [Yang and Newsam, 2010], ALOI-100 [Geusebroek et al., 2005], and AWA [Romera Paredes and Torr, 2015].
Dataset Splits No To evaluate the performance for incomplete multi-view data, we randomly set the instances with a certain ratio (i.e., [0.1, 0.3, 0.5, 0.7]) as the missing pairs. This describes the method for introducing incompleteness, but not the explicit training/test/validation splits for the datasets themselves.
Hardware Specification Yes For all experiments, we employ a Linux platform equipped with an NVIDIA RTX 4090 GPU and 32GB of memory
Software Dependencies Yes using Py Torch version 2.3.0.
Experiment Setup Yes To be specific, the view-specific encoder and decoder layers are configured with dimensions of (0.8dv, 0.8dv, 1500, C) and (C, 1500, 0.8dv, 0.8dv, dv), respectively. ... We set the temperature parameters to σ = 0.1 and θ = 0.05. ... the optimal value of λ,α and β, i.e. λ = 0.5, α=0.005 or 0.01, and β=0.005 or 0.01.