I would like to thank Jieyu Zhao (supported by SNF grant 21-43'417.95 ``Incremental Self-improvement''), Nic Schraudolph, Rafal Salustowicz, Sepp Hochreiter, Marco Wiering, and Luca Gambardella for valuable discussions and comments on a draft of this chapter, which is based on the following earlier publications: [28,30].