A Semi-supervised Generative Model for Incomplete Multi-view Data Integration with Missing Labels
This addresses the challenge of missing views and labels in multi-view learning, such as in biological data, but is incremental as it builds on prior probabilistic methods.
The paper tackles the problem of incomplete multi-view data integration with missing labels by proposing a semi-supervised generative model that leverages both labeled and unlabeled samples. It achieves better predictive and imputation performance on image and multi-omics data compared to existing approaches.
Multi-view learning is widely applied to real-life datasets, such as multiple omics biological data, but it often suffers from both missing views and missing labels. Prior probabilistic approaches addressed the missing view problem by using a product-of-experts scheme to aggregate representations from present views and achieved superior performance over deterministic classifiers, using the information bottleneck (IB) principle. However, the IB framework is inherently fully supervised and cannot leverage unlabeled data. In this work, we propose a semi-supervised generative model that utilizes both labeled and unlabeled samples in a unified framework. Our method maximizes the likelihood of unlabeled samples to learn a latent space shared with the IB on labeled data. We also perform cross-view mutual information maximization in the latent space to enhance the extraction of shared information across views. Compared to existing approaches, our model achieves better predictive and imputation performance on both image and multi-omics data with missing views and limited labeled samples.