Self-paced and soft-weighted nonnegative matrix factorization for data representation
Self-paced and soft-weighted nonnegative matrix factorization for data representation is a scholarly work, published in 2019 in ''Knowledge-Based Systems''. The main subjects of the publication include political representation, non-negative matrix factorization, convergence, feature engineering, artificial intelligence, computer science, matrix, weighting, mathematical optimization, matrix decomposition, scheme, facial recognition system, convexity, algorithm, and benchmark. To alleviate this deficiency, this paper presents a novel NMF method by gradually including data points into NMF from easy to complex, namely self-paced learning (SPL), which is shown to be beneficial in avoiding a bad local solution.