Reproduce “Correlated Feature Selection with Extended Exclusive Group Lasso” 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
Yuxin Sun, Benny Chain, Samuel Kaski, John Shawe-Taylor
The authors introduce a feature-selection method for correlated biological measurements with unknown groups. It randomly assigns features to groups, repeats selection across subsamples, and uses artificial informative features to control irrelevant selections. For CMV prediction, it selected 47 TCRs, all over-represented in CMV-positive samples, and achieved 0.88 and 0.89 accuracy in two held-out cohorts; this does not establish CMV specificity.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.