Cooperative Learning for Multi-view Analysis

The package multiview is a new method for supervised learning with multiple sets of features called views. The multi-view problem is especially important in biology and medicine, where “-omics” data such as genomics, proteomics and radiomics are measured on a common set of samples. Cooperative learning combines the usual squared error loss of predictions with an “agreement” penalty to encourage the predictions from different data views to agree. By varying the weight of the agreement penalty, we get a continuum of solutions that include the well-known early and late fusion approaches. Cooperative learning chooses the degree of agreement (or fusion) in an adaptive manner, using a validation set or cross-validation to estimate test set prediction error.

In the setting of cooperative regularized linear regression, the method combines the lasso penalty with the agreement penalty, yielding feature sparsity. The method can be especially powerful when the different data views share some underlying relationship in their signals that can be exploited to boost the signals.

As shown in Ding et al. (2021), cooperative learning achieves higher predictive accuracy on simulated data and real multiomics examples of labor onset prediction and breast ductal carcinoma in situ and invasive breast cancer classification. Leveraging aligned signals and allowing flexible fitting mechanisms for different modalities, cooperative learning offers a powerful approach to multiomics data fusion.

References

Ding, Daisy Yi, Shuangning Li, Balasubramanian Narasimhan, and Robert Tibshirani. 2021. “Cooperative Learning for Multi-View Analysis.” arXiv Preprint arXiv:2112.12337.