Building a robot that learns to perform a task has been acknowledged as one of the major challenges facing artificial intelligence. Self-improving robots would relieve humans from much of the drudgery of programming and would potentially allow operation in environments that were changeable or only partially known. Progress towards this goal would also make fundamental contributions to artificial intelligence by furthering our understanding of how to successfully integrate disparate abilities such as perception, planning, learning and action.
Although its roots can be traced back to the...

Learning Representation and Control in Markov Decision Processes describes methods for automatically compressing Markov decision processes (MDPs) by learning a low-dimensional linear approximation defined by an orthogonal set of basis functions. A unique feature of the text is the use of Laplacian operators, whose matrix representations have non-positive off-diagonal elements and zero row sums. The generalized inverses of Laplacian operators, in particular the Drazin inverse, are shown to be useful in the exact and approximate solution of MDPs. The author goes on to describe a broad framework...

Building a robot that learns to perform a task has been acknowledged as one of the major challenges facing artificial intelligence. Self-improving robots would relieve humans from much of the drudgery of programming and would potentially allow operation in environments that were changeable or only partially known. Progress towards this goal would also make fundamental contributions to artificial intelligence by furthering our understanding of how to successfully integrate disparate abilities such as perception, planning, learning and action.
Although its roots can be traced back to the...