Preconditioned Temporal Difference Learning

This paper extends many of the recent popular reinforcement learning (RL) algorithms to a generalized framework that includes least-squares temporal difference (LSTD) learning, least-squares policy evaluation (LSPE) and a variant of incremental LSTD (iLSTD). The basis of this extension is a preconditioning technique that tries to solve a stochastic model equation. This paper also studies three signicant issues of the new framework: it presents a new rule of step-size that can be computed online, provides an iterative way to apply preconditioning, and reduces the complexity of related algorithms to near that of temporal difference (TD) learning.