@inproceedings{d05b5812cad24d0d83d94073ca07186c,
title = "Optimistic AIXI",
abstract = "We consider extending the AIXI agent by using multiple (or even a compact class of) priors. This has the benefit of weakening the conditions on the true environment that we need to prove asymptotic optimality. Furthermore, it decreases the arbitrariness of picking the prior or reference machine. We connect this to removing symmetry between accepting and rejecting bets in the rationality axiomatization of AIXI and replacing it with optimism. Optimism is often used to encourage exploration in the more restrictive Markov Decision Process setting and it alleviates the problem that AIXI (with geometric discounting) stops exploring prematurely.",
keywords = "AIXI, Optimality, Optimism, Reinforcement Learning",
author = "Peter Sunehag and Marcus Hutter",
year = "2012",
doi = "10.1007/978-3-642-35506-6_32",
language = "English",
isbn = "9783642355059",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "312--321",
booktitle = "Artificial General Intelligence - 5th International Conference, AGI 2012, Proceedings",
note = "5th International Conference on Artificial General Intelligence, AGI 2012 ; Conference date: 08-12-2012 Through 11-12-2012",
}