English

Kryptonite-N: Machine Learning Strikes Back

Machine Learning 2025-01-28 v2 Artificial Intelligence

Abstract

Quinn et al propose challenge datasets in their work called ``Kryptonite-N". These datasets aim to counter the universal function approximation argument of machine learning, breaking the notation that machine learning can ``approximate any continuous function" \cite{original_paper}. Our work refutes this claim and shows that universal function approximations can be applied successfully; the Kryptonite datasets are constructed predictably, allowing logistic regression with sufficient polynomial expansion and L1 regularization to solve for any dimension N.

Cite

@article{arxiv.2412.20588,
  title  = {Kryptonite-N: Machine Learning Strikes Back},
  author = {Albus Li and Nathan Bailey and Will Sumerfield and Kira Kim},
  journal= {arXiv preprint arXiv:2412.20588},
  year   = {2025}
}
R2 v1 2026-06-28T20:51:26.992Z