Reproduce “Computing Functions of Random Variables via Reproducing Kernel Hilbert Space Representations” 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
Bernhard Schölkopf, Krikamol Muandet, Kenji Fukumizu, Jonas Peters
The authors propose kernel probabilistic programming (KPP), which represents distributions as kernel-feature-space means and applies operations to combinations of sampled points, without first estimating densities. In synthetic tests, error generally declined as sample size grew for multiplication, division, and exponentiation. On 81 CauseEffectPairs, KPP achieved 80% forced-decision causal-direction accuracy; this preliminary result has no reported baseline comparison.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.