Introduction to Reinforcement Learning
Introduction to Reinforcement Learning
Universal Artificial Intelligence: Sequential Decisions Based On Algorithmic Probability
Universal Artificial Intelligence: Sequential Decisions Based On Algorithmic Probability
An Introduction to Kolmogorov Complexity and Its Applications
An Introduction to Kolmogorov Complexity and Its Applications
Optimality issues of universal greedy agents with static priors
ALT'10 Proceedings of the 21st international conference on Algorithmic learning theory
A Monte-Carlo AIXI approximation
Journal of Artificial Intelligence Research
Self-modification and mortality in artificial agents
AGI'11 Proceedings of the 4th international conference on Artificial general intelligence
Delusion, survival, and intelligent agents
AGI'11 Proceedings of the 4th international conference on Artificial general intelligence
Universal knowledge-seeking agents
ALT'11 Proceedings of the 22nd international conference on Algorithmic learning theory
ALT'11 Proceedings of the 22nd international conference on Algorithmic learning theory
Complexity-based induction systems: Comparisons and convergence theorems
IEEE Transactions on Information Theory
Asymptotic non-learnability of universal agents with computable horizon functions
Theoretical Computer Science
| Hi-index | 5.23 |
Reinforcement learning (RL) agents like Hutter@?s universal, Pareto optimal, incomputable AIXI heavily rely on the definition of the rewards, which are necessarily given by some ''teacher'' to define the tasks to solve. Therefore, as is, AIXI cannot be said to be a fully autonomous agent. From the point of view of artificial general intelligence (AGI), this can be argued to be an incomplete definition of a generally intelligent agent. Furthermore, it has recently been shown that AIXI can converge to a suboptimal behavior in certain situations, hence showing the intrinsic difficulty of RL, with its non-obvious pitfalls. We propose a new model of intelligence, the knowledge-seeking agent (KSA), halfway between Solomonoff induction and AIXI, that defines a completely autonomous agent that does not require a teacher. The goal of this agent is not to maximize arbitrary rewards, but to entirely explore its world in an optimal way. A proof of strong asymptotic optimality for a class of horizon functions shows that this agent behaves according to expectation. Some implications of such an unusual agent are proposed.