Elastic-Net Multiple Kernel Learning: Combining Multiple Data Sources for Prediction

A novel approach to core set selection in machine learning has been introduced, utilizing approval-based multi-winner voting where instances act as both voters and candidates. This method defines the approval set of each instance based on existing literature and selects winners through a representative voting rule, leading to a reduced training set. Experimental results indicate significant performance improvements over state-of-the-art methods using neural network classifiers and traditional classifiers like KNN and SVM.

A novel method called STARK has been developed to denoise spatial transcriptomics images, enhancing the identification of cell types at ultra-low sequencing depths. This technique employs adaptive regularization and kernel ridge regression, updating a spatial graph iteratively to improve image quality and gene expression interpolation.