Solomonoff induction violates Nicod's criterion

DOI10.1007/978-3-319-24486-0_23zbMATH Open1388.68142DBLPconf/alt/LeikeH15arXiv1507.04121OpenAlexW2963361331WikidataQ58012203 ScholiaQ58012203MaRDI QIDQ2835642FDOQ2835642

Authors: Jan Leike, Marcus Hutter 

Publication date: 30 November 2016

Published in: Lecture Notes in Computer Science (Search for Journal in Brave)

Abstract: Nicod's criterion states that observing a black raven is evidence for the hypothesis H that all ravens are black. We show that Solomonoff induction does not satisfy Nicod's criterion: there are time steps in which observing black ravens decreases the belief in H. Moreover, while observing any computable infinite string compatible with H, the belief in H decreases infinitely often when using the unnormalized Solomonoff prior, but only finitely often when using the normalized Solomonoff prior. We argue that the fault is not with Solomonoff induction; instead we should reject Nicod's criterion.

Full work available at URL: https://arxiv.org/abs/1507.04121

Recommendations

• Natural descriptions and anthropic bias: extant problems in Solomonoff induction
• Modes of convergence to the truth: steps toward a better epistemology of induction
• On universal prediction and Bayesian confirmation
• Theory and Applications of Models of Computation
• Principles of Solomonoff Induction and AIXI

zbMATH Keywords

confirmationequivalence conditionBayesian reasoningdisconfirmationHempel's paradoxSolomonoff normalization

Mathematics Subject Classification ID

Probability and inductive logic (03B48) Algorithmic information theory (Kolmogorov complexity, etc.) (68Q30)

Cites Work

• Merging of Opinions with Increasing Information
• Probability Theory
• A formal theory of inductive inference. Part I
• Title not available (Why is that?)
• An introduction to Kolmogorov complexity and its applications
• On the relation between descriptional complexity and algorithmic probability
• Complexity-based induction systems: Comparisons and convergence theorems
• I.—STUDIES IN THE LOGIC OF CONFIRMATION (I.)
• On universal prediction and Bayesian confirmation
• A philosophical treatise of universal induction
• Hempel's Raven Paradox: A Lacuna in the Standard Bayesian Solution
• (Non-)Equivalence of Universal Priors
• New error bounds for Solomonoff prediction
• The generalized radially symmetric wave equation

This page was built for publication: Solomonoff induction violates Nicod's criterion

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2835642)