Extending Universal Intelligence Models with Formal Notion of Representation
Abstract
Solomonoff induction is known to be universal, but incomputable. Its approximations, namely, the Minimum Description (or Message) Length (MDL) principles, are adopted in practice in the efficient, but non-universal form. Recent attempts to bridge this gap leaded to development of the Representational MDL principle that originates from formal decomposition of the task of induction. In this paper, possible extension of the RMDL principle in the context of universal intelligence agents is considered, for which introduction of representations is shown to be an unavoidable meta-heuristic and a step toward efficient general intelligence. Hierarchical representations and model optimization with the use of information-theoretic interpretation of the adaptive resonance are also discussed.
Keywords
Cite
@article{arxiv.1306.1557,
title = {Extending Universal Intelligence Models with Formal Notion of Representation},
author = {Alexey Potapov and Sergey Rodionov},
journal= {arXiv preprint arXiv:1306.1557},
year = {2013}
}
Comments
proceedings of AGI 2012, Lecture Notes in Artificial Intelligence, Vol. 7716, pp. 242-251, Springer-Verlag, 2012. The final publication is available at link.springer.com