How Gruesome are the No-free-lunch Theorems for Machine Learning?

Croatian Journal of Philosophy 18 (3):479-485 (2018)
  Copy   BIBTEX

Abstract

No-free-lunch theorems are important theoretical result in the fields of machine learning and artificial intelligence. Researchers in this fields often claim that the theorems are based on Hume’s argument about induction and represent a formalisation of the argument. This paper argues that this is erroneous but that the theorems correspond to and formalise Goodman’s new riddle of induction. To demonstrate the correspondence among the theorems and Goodman’s argument, a formalisation of the latter in the spirit of the former is sketched.

Links

PhilArchive



    Upload a copy of this work     Papers currently archived: 92,953

External links

Setup an account with your affiliations in order to access resources via your University's proxy server

Through your library

Similar books and articles

The Implications of the No-Free-Lunch Theorems for Meta-induction.David H. Wolpert - 2023 - Journal for General Philosophy of Science / Zeitschrift für Allgemeine Wissenschaftstheorie 54 (3):421-432.
Intelligent design and the NFL theorems.Olle Häggström - 2007 - Biology and Philosophy 22 (2):217-230.
Simulation of biological evolution and the nfl theorems.Ronald Meester - 2009 - Biology and Philosophy 24 (4):461-472.
Philosophy through Machine Learning.Daniel Lim - 2020 - Teaching Philosophy 43 (1):29-46.
No Free Lunch Theorems for Optimization.D. H. Wolpert & W. G. Macready - 1997 - IEEE Transactions on Evolutionary Computation 1 (1):67–82.
Gödel’s Incompleteness Theorems and Artificial Life.John P. Sullins - 1997 - Society for Philosophy and Technology Quarterly Electronic Journal 2 (3):185-195.

Analytics

Added to PP
2019-02-15

Downloads
22 (#731,954)

6 months
3 (#1,045,430)

Historical graph of downloads
How can I increase my downloads?

Author's Profile

Davor Lauc
University of Zagreb

References found in this work

No references found.

Add more references