pith. sign in

arxiv: 0707.3087 · v3 · submitted 2007-07-20 · 💻 cs.IT · cs.LG· math.IT

Universal Reinforcement Learning

classification 💻 cs.IT cs.LGmath.IT
keywords algorithmcostactionsactiveagentaveragefutureobservations
0
0 comments X
read the original abstract

We consider an agent interacting with an unmodeled environment. At each time, the agent makes an observation, takes an action, and incurs a cost. Its actions can influence future observations and costs. The goal is to minimize the long-term average cost. We propose a novel algorithm, known as the active LZ algorithm, for optimal control based on ideas from the Lempel-Ziv scheme for universal data compression and prediction. We establish that, under the active LZ algorithm, if there exists an integer $K$ such that the future is conditionally independent of the past given a window of $K$ consecutive actions and observations, then the average cost converges to the optimum. Experimental results involving the game of Rock-Paper-Scissors illustrate merits of the algorithm.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.