Reproduce “Learning Non-Linear Feature Maps” 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
Dimitrios Athanasakis, John Shawe-Taylor, Delmiro Fernandez-Reyes
randSel selects features by comparing kernel–label alignment across matched random subsets with and without each candidate, then repeatedly removes the lowest-scoring quarter. In a 100-feature XOR benchmark, it selected 2.0 ± 0.0 features with 95.7 ± 3.3% accuracy and perfect precision and recall, tying backward-greedy BaHsic; this contrasts with forward-greedy FoHsic, not all alternatives.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.