Reproduce “Scalable Logistic Gaussian Process Density Regression with Kinetic Langevin Sampling” 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
Daniel Paulin, Ádám Jung, András A. Benczúr
The authors estimate full redshift distributions for galaxies rather than single photometric-redshift values. Their logistic Gaussian process uses Fourier and Nyström features with minibatch kinetic-Langevin sampling to integrate over density uncertainty. On DESI DR1 BGS, mean test NLL is −2.3786 versus −2.3564 for TabPFN-3.5, using 3.91 million versus 500,000 training rows; performance weakens under covariate shift.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.