K$ PCA

K-principal component analysis (k-PCA) aims to identify the top *k* principal components of a dataset, effectively reducing dimensionality while preserving maximal variance. Recent research focuses on improving the efficiency and robustness of k-PCA algorithms, particularly exploring deflation methods which iteratively find principal components. These advancements yield improved sample complexity and robustness to noisy or contaminated data, enhancing the applicability of k-PCA in various data analysis and machine learning tasks.