Hostname: page-component-586b7cd67f-dlnhk Total loading time: 0 Render date: 2024-11-23T22:10:34.905Z Has data issue: false hasContentIssue false
  • English
  • Français

No Free Lunch Theorem, Inductive Skepticism, and the Optimality of Meta-induction

Published online by Cambridge University Press:  01 January 2022

Gerhard Schurz
Get access Rights & Permissions [Opens in a new window]

Abstract

The no free lunch theorem is a radicalized version of Hume’s induction skepticism. It asserts that relative to a uniform probability distribution over all possible worlds, all computable prediction algorithms—whether ‘clever’ inductive or ‘stupid’ guessing methods (etc.)—have the same expected predictive success. This theorem seems to be in conflict with results about meta-induction. According to these results, certain meta-inductive prediction strategies may dominate other (non-meta-inductive) methods in their predictive success (in the long run). In this article this conflict is analyzed and dissolved, by means of probabilistic analysis and computer simulation.

Type
Evidence and Inference
Information
Philosophy of Science , Volume 84 , Issue 5 , December 2017 , pp. 825 - 839
Copyright
Copyright © The Philosophy of Science Association

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

†

This work was supported by the DFG (Deutsche Forschungsgemeinschaft), SPP 1516. For valuable help I am indebted to Paul Thorn and Ronald Ortner.

References

Carnap, Rudolf. 1950. Logical Foundations of Probability. Chicago: University of Chicago Press.Google Scholar
Cesa-Bianchi, Nicolo, and Lugosi, Gabor. 2006. Prediction, Learning, and Games. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
Earman, John. 1992. Bayes or Bust? Cambridge, MA: MIT Press.Google Scholar
Howson, Colin, and Urbach, Peter. 1996. Scientific Reasoning: The Bayesian Approach. 2nd ed. Chicago: Open Court.Google Scholar
Rao, R. Bharat, Gordon, Diana, and Spears, William. 1995. “For Every Generalization Action, Is There Really an Equal and Opposite Reaction?” In Machine Learning: Proceedings of ICML 1995, ed. Preditis, A. and Russell, S. J., 471–79. Burlington, MA: Morgan Kaufmann.Google Scholar
Reichenbach, Hans. 1949. The Theory of Probability. Berkeley: University of California Press.Google Scholar
Schaffer, Cullen. 1994. “A Conservation Law for Generalization Performance.” In Machine Learning: Proceedings of ICML 1994, ed. Cohen, W. W. and Hirsh, H., 259–65. Burlington, MA: Morgan Kaufmann.Google Scholar
Schurz, Gerhard. 2008. “The Meta-inductivist’s Winning Strategy in the Prediction Game: A New Approach to Hume’s Problem.” Philosophy of Science 75:278–305.CrossRefGoogle Scholar
Schurz, Gerhard, and Thorn, Paul. 2016. “The Revenge of Ecological Rationality: Strategy-Selection by Meta-induction.” Minds and Machines 26 (1): 31–59.CrossRefGoogle Scholar
Skyrms, Brian. 2000. Choice and Chance. 4th ed. Belmont, CA: Wadsworth.Google Scholar
Solomonoff, Ray J. 1964. “A Formal Theory of Inductive Inference.” Pts. 1 and 2. Information and Control 7:1–22, 224–54.CrossRefGoogle Scholar
Wolpert, David H. 1996. “The Lack of A Priori Distinctions between Learning Algorithms.” Neural Computation 8 (7): 1341–90.Google Scholar