Robust Multi-view Clustering against Imperfect Information
2026-06-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address a problem in multi-view clustering where data from different views is often incomplete or incorrectly matched. They propose a new method called Posterior-guided Latent Counterpart Inference (PLCI) that treats the matching counterpart of each data point in another view as a hidden variable to be inferred. PLCI uses information about both how reliable individual instances are and semantic grouping to improve these inferences. Tests on six datasets show that PLCI performs better than other recent methods at handling missing and noisy data. The authors plan to share their code once their work is accepted.
multi-view clusteringincomplete viewsnoisy correspondenceslatent variableposterior distributionsemantic transportinstance-level reliabilitycross-view matchingdata imputationrobust clustering
Authors
Zhichao Huang, Haochen Zhou, Hao Wang, Mouxing Yang, Xi Peng
Abstract
Real-world multi-view data always suffer from imperfect information problem, where the view-specific observations are absent (i.e., Incomplete Views, IV) and cross-view correspondences are mismatched (i.e., Noisy Correspondences, NC) for certain instances. As a remedy, numerous IV- and NC-oriented multi-view clustering (MvC) methods have been proposed, which however require either reliable correspondences or sufficiently complete instances, thus stopping short of addressing the imperfect information problem. In contrast, we observe that both IV and NC challenges originate from the same issue of imperfect cross-view counterpart information, where the counterpart of an anchor instance in another view might be either unavailable or unreliable. Based on the observation, we propose a novel robust MvC framework, termed Posterior-guided Latent Counterpart Inference (PLCI), which could handle both IV and NC in a unified manner. Specifically, PLCI formulates the desired cross-view counterpart of each anchor instance as a latent variable, and integrates both instance-level reliability and prototype-level semantic transport to infer the posterior distribution of the latent counterpart. Extensive experiments on six widely-used multi-view datasets against 10 state-of-the-art MvC methods demonstrate the effectiveness of PLCI for tackling the imperfect information problem. The code will be released upon acceptance.