The field of high-dimensional data analysis is moving towards developing more efficient and effective methods for feature selection, scaling, and variable selection. Researchers are focusing on improving the performance of machine learning algorithms by identifying and selecting the most relevant features, as well as developing scalable methods for handling large datasets. The use of novel initialization strategies, such as the Minkowski weighted $k$-means++, and the development of safe screening rules for group SLOPE, are enabling significant gains in computational efficiency and memory usage. Furthermore, the application of low-rank factorizations of conditional correlation matrices in graph learning is allowing for more efficient dimension-versus-performance trade-offs. Noteworthy papers include:

• A study on the impact of feature scaling in machine learning, which provides comprehensive guidance on the selection of feature scaling techniques for various machine learning algorithms.
• A proposal of a safe screening rule for Group SLOPE, which achieves considerable gains in computational efficiency and memory usage without compromising accuracy.