A Monte Carlo Algorithm for Universally Optimal Bayesian Sequence Prediction and Planning

Details A Monte Carlo Algorithm for Universally Optimal Bayesian Sequence Prediction and Planning A Monte Carlo Algorithm for Universally Optimal Bayesian Sequence Prediction and Planning

2010-01-18

arxiv.org/abs/1001.2813v1

Nonlinear Sciences

Adaptation and Self-Organizing Systems

Submitted to MDPI Algorithms Special Issue "Algorithmic Complexity in Physics & Embedded Artificial Intelligences"

Scientific paper

The aim of this work is to address the question of whether we can in principle design rational decision-making agents or artificial intelligences embedded in computable physics such that their decisions are optimal in reasonable mathematical senses. Recent developments in rare event probability estimation, recursive bayesian inference, neural networks, and probabilistic planning are sufficient to explicitly approximate reinforcement learners of the AIXI style with non-trivial model classes (here, the class of resource-bounded Turing machines). Consideration of the effects of resource limitations in a concrete implementation leads to insights about possible architectures for learning systems using optimal decision makers as components.

Affiliated with

Also associated with

No associations

Rating

A Monte Carlo Algorithm for Universally Optimal Bayesian Sequence Prediction and Planning does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with A Monte Carlo Algorithm for Universally Optimal Bayesian Sequence Prediction and Planning, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A Monte Carlo Algorithm for Universally Optimal Bayesian Sequence Prediction and Planning will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-659410