Asymptotically Optimal Agents
Artificial Intelligence
2012-02-10 v1 Machine Learning
Abstract
Artificial general intelligence aims to create agents capable of learning to solve arbitrary interesting problems. We define two versions of asymptotic optimality and prove that no agent can satisfy the strong version while in some cases, depending on discounting, there does exist a non-computable weak asymptotically optimal agent.
Cite
@article{arxiv.1107.5537,
title = {Asymptotically Optimal Agents},
author = {Tor Lattimore and Marcus Hutter},
journal= {arXiv preprint arXiv:1107.5537},
year = {2012}
}
Comments
21 LaTeX pages