Reproduce “Adaptive Uncertainty Quantification on Simplicial Complexes with Conformalized Simplicial Neural Networks” 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

Lekun Wang, Jaidev Goel, Yulia R. Gel, Yao Xie

CF-SNN wraps conformal prediction around a simplicial neural network (SNN) to produce sets for higher-order labels. It calibrates scores on labeled targets while allowing the model to use topology and features, under random splits and permutation-invariant scores. In citation-count regression, it reports 0.902 coverage and 13.24 interval width; simultaneous guarantees require no tied scores and may yield full response spaces.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.