Asymptotically Optimal Agents

ALT 2011 pp. 368-382

doi:10.1007/978-3-642-24412-4_29 /alt/2011/lattimore2011alt-asymptotically/

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.

PDF ALT Semantic Scholar

Cite

Text

Lattimore and Hutter. "Asymptotically Optimal Agents." International Conference on Algorithmic Learning Theory, 2011. doi:10.1007/978-3-642-24412-4_29

Markdown

[Lattimore and Hutter. "Asymptotically Optimal Agents." International Conference on Algorithmic Learning Theory, 2011.](https://mlanthology.org/alt/2011/lattimore2011alt-asymptotically/) doi:10.1007/978-3-642-24412-4_29

BibTeX

@inproceedings{lattimore2011alt-asymptotically,
  title     = {{Asymptotically Optimal Agents}},
  author    = {Lattimore, Tor and Hutter, Marcus},
  booktitle = {International Conference on Algorithmic Learning Theory},
  year      = {2011},
  pages     = {368-382},
  doi       = {10.1007/978-3-642-24412-4_29},
  url       = {https://mlanthology.org/alt/2011/lattimore2011alt-asymptotically/}
}