This paper proposes a novel research direction for algorithmic probability theory, which now allows to formalize philosophical questions about concepts such as the simulation argument, personal identity, and the rationality of utilitarianism. Pushing progress in these directions might be of high relevance for future society and artificial general intelligence. We start to build up a set of formulae while relating them to the real world. We then use our gained understanding to outline in a few examples how to correctly decide between conflicting philosophical hypotheses. Our approach arises from combining the essential ideas of formalisms such as: Universal intelligence measure, space-time-embedded intelligence, Hutter observer localisation.
from cs.AI updates on arXiv.org http://ift.tt/1iqPl8U
via IFTTT
No comments:
Post a Comment