In abstraction selection, an agent must choose an abstraction from a set of candidate abstractions, each build up from a different combination of state ...

We examine our approach in three domains of increasing complexity: contextual bandit problems, episodic MDPs, and general MDPs with context- specific structure.

Aug 6, 2013 · This article addresses reinforcement learning problems based on factored Markov decision processes (MDPs) in which the agent must choose ...

The core of the approach is to make selection of an abstraction part of the learning agent's decision-making process by augmenting the agent's action space with ...

PDF | This paper introduces a novel approach for abstraction selec-tion in reinforcement learning problems modelled as factored Markov decision.

This paper introduces a novel approach for abstraction selection in reinforcement learning problems modelled as factored Markov decision processes (MDPs), ...

Jul 4, 2013 · In abstraction selection, an agent must choose an abstraction from a set of candidate abstractions, each build up from a different combination ...

We present an abstraction selection algorithm, show empirically that it selects an appropriate abstraction us- ing very little sample data, and that it ...

State abstractions are often used to reduce the complexity of model-based reinforcement learn- ing when only limited quantities of data are avail- able. ...

This paper introduces a novel approach for abstraction selection in reinforcement learning problems modelled as factored Markov decision processes (MDPs), ...