Technical Note: \cal Q-Learning
Machine Learning
Introduction to Reinforcement Learning
Introduction to Reinforcement Learning
Journal of Logic, Language and Information
PAC model-free reinforcement learning
ICML '06 Proceedings of the 23rd international conference on Machine learning
Universal Intelligence: A Definition of Machine Intelligence
Minds and Machines
An Introduction to Kolmogorov Complexity and Its Applications
An Introduction to Kolmogorov Complexity and Its Applications
Learning Mazes with Aliasing States: An LCS Algorithm with Associative Perception
Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
A universal measure of intelligence for artificial agents
IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
Measuring universal intelligence: Towards an anytime intelligence test
Artificial Intelligence
Environments for multiagent systems state-of-the-art and research challenges
E4MAS'04 Proceedings of the First international conference on Environments for Multi-Agent Systems
On Potential Cognitive Abilities in the Machine Kingdom
Minds and Machines
| Hi-index | 0.00 |
In this paper we apply the recent notion of anytime universal intelligence tests to the evaluation of a popular reinforcement learning algorithm, Q-learning. We show that a general approach to intelligence evaluation of AI algorithms is feasible. This top-down (theory-derived) approach is based on a generation of environments under a Solomonoff universal distribution instead of using a pre-defined set of specific tasks, such as mazes, problem repositories, etc. This first application of a general intelligence test to a reinforcement learning algorithm brings us to the issue of task-specific vs. general AI agents. This, in turn, suggests new avenues for AI agent evaluation and AI competitions, and also conveys some further insights about the performance of specific algorithms.