Physical Sciences and Mathematics Commons^™

Consensus Regularized Selection Based Prediction, Ping Wang

Wayne State University Theses

Integrating regularization methods within a regression framework has become a popular choice for researchers to build predictive models with lower variance and better generalization. Regularizers also aid in building interpretable models with high-dimensional data which makes them very appealing. Regularizers in general are unique in nature as they cater to data specific features such as correlation, structured sparsity, and temporal smoothness. The problem of obtaining a consensus among such diverse regularizers is extremely important in order to determine the optimal regularizer for the model. This is called the consensus regularization problem which has not received much attention in the literature, …