Policy gradient (PG) reinforcement learning algorithms have strong (local) convergence guarantees, but their learning performance is typically limited by a large variance in the e...
T ORDER REGRESSION (EXTENDED ABSTRACT) Kurt Driessensa Saso Dzeroskib a Department of Computer Science, University of Waikato, Hamilton, New Zealand (kurtd@waikato.ac.nz) b Departm...
Basically, instrumental conditioning is learning through consequences: Behavior that produces positive results (high “instrumental response”) is reinforced, and that which pro...
We present a reinforcement learning architecture, Dyna-2, that encompasses both samplebased learning and sample-based search, and that generalises across states during both learni...
In this paper, we show how adaptive prototype optimization can be used to improve the performance of function approximation based on Kanerva Coding when solving largescale instanc...