AI Paper Atlas

Research map / Learning theory and algorithms / Clustering

Semantic Invariant Multi-View Clustering With Fully Incomplete Information

Pengxin Zeng, Mouxing Yang, Yiding Lu, Changqing Zhang, Peng Hu, Xi Peng

TPAMI 2023 · Learning theory and algorithms › Clustering · approach: Incomplete multi-view clustering · 70 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the cross-view term I(C;V) (L_SIL-v) to reconstruction (ablation)SMILE · NoisyMNIST (100% unaligned)clustering ACC (%): 50.1 -> 74.5Improved
add the cross-view term I(C;V) (L_SIL-v) to reconstruction (ablation)SMILE · NoisyMNIST (100% missing)clustering ACC (%): 48.7 -> 66.8Improved
add the semantic discrimination term I(C;X) (L_SIL-s) to reconstruction (ablation)SMILE · NoisyMNIST (100% unaligned)clustering ACC (%): 50.1 -> 52.7Improved
add the semantic discrimination term I(C;X) (L_SIL-s) to reconstruction (ablation)SMILE · NoisyMNIST (100% missing)clustering ACC (%): 48.7 -> 51.0Improved
add L_SIL-s on top of L_SIL-v (ablation)SMILE · NoisyMNIST (100% unaligned)clustering ACC (%): 74.5 -> 78.2Improved
add L_SIL-s on top of L_SIL-v (ablation)SMILE · NoisyMNIST (100% missing)clustering ACC (%): 66.8 -> 67.8Improved
replace plain reconstruction with discrepancy-aware reconstruction (DAR) (ablation)SMILE · NoisyMNIST (100% unaligned)clustering ACC (%): 78.2 -> 82.7Improved
replace plain reconstruction with discrepancy-aware reconstruction (DAR) (ablation)SMILE · NoisyMNIST (100% missing)clustering ACC (%): 67.8 -> 69.0Improved
set gamma weight of cross-view term to 0 (ablation)SMILE · MNISTUSPS and NoisyMNIST (100% unaligned/missing/unpaired)clustering ACC (%)Hurt
use SMILE instead of MVC-UM when all correspondences are unalignedSMILE · NoisyMNIST (100% unaligned)clustering ACC (%): 19.3 -> 82.7Improved
use SMILE instead of DM2C when all instances are missingSMILE · NoisyMNIST (100% missing)clustering ACC (%): 23.2 -> 69.0Improved
increase lambda_SIL for imputationSMILE · Caltech imputationNRMSEImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy