Reproduce “An analysis of NK and generalized NK landscapes” 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
Jeffrey S. Buzas, Jeffrey Dinitz
The authors recast the classic NK landscape as a statistical model of main effects and interactions, making model rank and expected local-optimum counts analyzable through multivariate-normal orthant probabilities. For fixed N,K designs, higher rank is associated with fewer expected optima, but generalized landscapes show the opposite, and peak count imperfectly indicates search difficulty.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.