Reproduce “Differential rank preservation manifold learning” 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
Henry Han, Luyu Niu, Diane Li, Jiacun Wang
The authors introduce Smooth Rank t-SNE (SR-tSNE), which adds a differentiable ranking loss to t-SNE’s similarity objective to preserve the order of original-space neighbors and rank them ahead of intruders. At k=10, mean Ordinal Degree of Locality Preservation across eight datasets is 0.0938, versus 0.0788 for t-SNE and 0.0792 for an equal-budget control; tests cover two-dimensional embeddings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.