The standard reinforcement learning view of the involvement of neuromodulatory systems in instrumental conditioning includes a rather straightforward conception of motivation as prediction of sum future reward. Competition between actions is based on the motivating characteristics of their consequent states in this sense. Substantial, careful, experiments reviewed in Dickinson & Balleine, into the neurobiology and psychology of motivation shows that this view is incomplete. In many cases, animals are faced with the choice not between many different actions at a given state, but rather whether a single response is worth executing at all. Evidence suggests that the motivational process underlying this choice has different psychological and neural properties from that underlying action choice. We describe and model these motivational systems, and consider the way they interact.

We present HEMP - a novel learning algorithm designed to operate in the domain of Markov processes and Markov decision processes. HEMP learns to perform a ...

We describe a novel algorithm that learns to perform a Heuristic Embedding of Markov Processes (HEMP) into a low dimensional Euclidean space (Engel & Mannor, 2001) Learning is performed online by observing actual state transitions and gradually constructing a map of the Markov state space.