Estimation and Approximation Bounds for Gradient-Based Reinforcement Learning.

Peter L. Bartlett Jonathan Baxter Estimation and Approximation Bounds for Gradient-Based Reinforcement Learning. 133-150 2002 64 J. Comput. Syst. Sci. 1 http://dx.doi.org/10.1006/jcss.2001.1793 db/journals/jcss/jcss64.html#BartlettB02