Reproduce “Classification EM-PCA for clustering and embedding” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zineddine Tighidet, Lazhar Labiod, Mohamed Nadif
CEM-PCA learns a low-dimensional representation and clusters together, rather than fixing the representation before clustering. It alternates between updating the embedding and fitting clusters. On Chang’s simulated 1,000-by-15 dataset with two classes, it reports 100% accuracy, versus 78% for K-means on PCA components. This comparison changes both representation and clustering method, so it does not isolate joint optimization as the cause.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.