@inproceedings{342cd0096e2044d789ce6cd7daa48b9e,
title = "Solomonoff induction violates nicod{\textquoteright}s criterion",
abstract = "Nicod{\textquoteright}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{\textquoteright}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{\textquoteright}s criterion.",
keywords = "Bayesian reasoning, Confirmation, Disconfirmation, Equivalence condition, Hempel{\textquoteright}s paradox, Solomonoff normalization",
author = "Jan Leike and Marcus Hutter",
note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 26th International Conference on Algorithmic Learning Theory, ALT 2015 ; Conference date: 04-10-2015 Through 06-10-2015",
year = "2015",
doi = "10.1007/978-3-319-24486-0_23",
language = "English",
isbn = "9783319244853",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "349--363",
editor = "Claudio Gentile and Sandra Zilles and Kamalika Chaudhuri",
booktitle = "Algorithmic Learning Theory - 26th International Conference, ALT 2015",
address = "Germany",
}