Reproduce “Gradient Boosting with Kolmogorov-Arnold Units for Interpretable Regression beyond Pairwise Interactions” 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

Minjong Cheon, Hanseon Joo, Eunji Lee

MJBoost models regression by boosting units that transform each input with a one-dimensional spline, sum the results, and apply a learned outer spline. This lets units represent higher-order interactions, which the authors separate from main and pair effects using functional ANOVA. On airfoil self-noise, test RMSE is 1.10 dB versus 3.41 for its order-≤2 truncation; this is one example.

Summary by alphaXiv.

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