REACTIVE LAYER IN AGI AGENT - Implementation of Adaptive Reactive Behavior and Beyond

Vilem Benes

2011

Abstract

Basic mechanisms of cognition working in AGI agent are presented. I argue that reactive behavior is the baseline of intelligence – it is the base component working and it can be further extended to produce more intelligent agents. Mechanisms employed at reactive level enable the agent to develop behavior which both explores and exploits the environment with the purpose of receiving highest reward possible. Three funda-mental mechanisms are intertwined – action selection, action value estimation and situation discrimination. Whole process of adaptation is completely unsupervised and depends only on reward received from envi-ronment. Some technical details of implementation of given mechanisms (BAGIB agent) are described to-gether with implications to other planned parts of “AGI-compliant” architecture. Discussed are several chal-lenges we encounter in AGI, which are not present in usually narrow and domain-limited approach to AI.

References

  1. Barto, A. G., Sutton, R. S. and Anderson C. W., 1983. Neuronlike elements that can solve difficult learning control problems, IEEE Transactions on Systems, Man, and Cybernetics, 13: 835-846.
  2. Benes, V., 2004. Intelligence means to be rewarded, Proceedings of the 5th International PhD. Workshop on Systems and Control. A Young Generation Viewpoint. Balatonfüred, Hungary. (online: http://home.zcu.cz/ shodan/).
  3. Benes, V., 2005. Defining Situations - Fundamental Level of Perception, Proceedings of the 6th International PhD Workshop on Systems and Control Young Generation Viewpoint. Ljubljana: Jozef Stefan Institute. (online: http://home.zcu.cz/shodan/).
  4. Dreyfus, H. L., 2007. Why Heideggerian AI Failed and How Fixing it Would Require Making it More Heideggerian, Philosophical Psychology, Volume 20, Issue 2 (April).
  5. Goertzel, B., 2008. AI and AGI: Past, Present and Future, AGI-08 Conference. University of Memphis. (online: http://www.acceleratingfuture.com/peopleblog/?p=1814).
  6. Legg, S. and Hutter, M., 2007. A collection of definitions of intelligence, Advances in Artificial General Intelligence: Concepts, Architectures and Algorithms. NL: IOS Press. (online: http://www.vetta.org/definitionsof-intelligence/).
  7. Minsky, M., 1981. Jokes and the Cognitive Unconscious, In Cognitive Constraints on Communication, Vaina and Hintikka (eds.). Reidel. (online: http://web.media. mit.edu/minsky/papers/jokes.cognitive.txt).
  8. Sutton, R. S. and Barto, A. G., 1998. Reinforcement Learning: An Introduction, MIT Press. (online: http://webdocs.cs.ualberta.ca/sutton/book/ebook/thebook.html).
Download


Paper Citation


in Harvard Style

Benes V. (2011). REACTIVE LAYER IN AGI AGENT - Implementation of Adaptive Reactive Behavior and Beyond . In Proceedings of the 3rd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART, ISBN 978-989-8425-40-9, pages 721-726. DOI: 10.5220/0003296607210726


in Bibtex Style

@conference{icaart11,
author={Vilem Benes},
title={REACTIVE LAYER IN AGI AGENT - Implementation of Adaptive Reactive Behavior and Beyond},
booktitle={Proceedings of the 3rd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,},
year={2011},
pages={721-726},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003296607210726},
isbn={978-989-8425-40-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 3rd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,
TI - REACTIVE LAYER IN AGI AGENT - Implementation of Adaptive Reactive Behavior and Beyond
SN - 978-989-8425-40-9
AU - Benes V.
PY - 2011
SP - 721
EP - 726
DO - 10.5220/0003296607210726