Abstract
Group lasso is a regularization widely used for feature group selection with sparsity at a group level in machine learning. Training a model with the group lasso regularization, however, leads to the selection of all the groups together that are closely related to each other although their features are useful to predict a target. In this study, we propose a new regularization, group-exclusive group lasso, for automatic exclusive feature group selection. The proposed regularization aims to enforce exclusive sparsity at an inter-group level, discouraging the coincident selection of the feature groups that are group-level correlated and share predictive powers toward the targets. The proposed method aims at higher group sparsity for selecting salient feature groups only, and is applied to neural networks. We evaluate the proposed regularization in neural networks on synthetic datasets and a real-life case for virtual metrology with automatic sensor selection in semiconductor manufacturing.
Original language | English |
---|---|
Pages (from-to) | 505-517 |
Number of pages | 13 |
Journal | IEEE Transactions on Semiconductor Manufacturing |
Volume | 37 |
Issue number | 4 |
DOIs | |
State | Published - 2024 |
Keywords
- Group exclusivity
- group sparsity
- regularization
- sensor selection
- virtual metrology