AutoNumerics-Zero: Automated Discovery of State-of-the-Art Mathematical Functions

Created by W.Langdon from gp-bibliography.bib Revision:1.8972

@InProceedings{Real:2026:ICML,

• author = "Esteban Real and Mirko Rossini and Connal {de Souza} and Manav Garg and Moritz Firsching and Quoc V. Le and Yao Chen and Akhil Verghese and Ekin Dogus Cubuk and David H. Park",
• title = "{AutoNumerics-Zero}: Automated Discovery of State-of-the-Art Mathematical Functions",
• booktitle = "Proceedings of the 43rd International Conference on Machine Learning",
• year = "2026",
• editor = "Alekh Agarwal and Miro Dudik and Martin Jaggi and Sharon Li",
• address = "Seoul, South Korea",
• month = jul # " 6-11",
• note = "Forthcoming",
• keywords = "genetic algorithms, genetic programming, symbolic regression, Machine Learning, cs.LG, Numerical Analysis, math.NA, NSGA-II, CMA-ES, JAX",
• URL = "https://icml.cc/virtual/2026/poster/61770",
• URL = "https://arxiv.org/abs/2312.08472",
• code_url = "https://github.com/google-deepmind/autonumerics_zero",
• size = "20 pages",
• abstract = ".. we discovered a 10-operation program that approximates the exponential function to 14 significant figures, exceeding the accuracy of previously known approximations of this size by more than 6 orders of magnitude.",
• notes = "Entered 2026 Humies.

  See also \cite{ereal2024autonumerics} https://arxiv.org/abs/2312.08472

  'ground truth was computed in float80'

  ICML 2026

  Google DeepMind",

}

Genetic Programming entries for Esteban Real Mirko Rossini Connal de Souza Manav Garg Moritz Firsching Quoc V Le Yao Chen Akhil Verghese Ekin Dogus Cubuk David H Park

Citations