×
Abstract. We propose a framework for approximate resolution of. MDPs with factored state space, factored action space and additive.
A framework for approximate resolution of MDPs with factored state space, factored action space and additive reward is proposed, based on considering ...
We propose a framework for generating a control policy for a traffic network of signalized intersections to accomplish control objectives expressed in linear ...
Abstract. We propose a framework for approximate resolution of MDPs with factored state space, factored action space and additive reward, ...
We describe some relevant background in planning, Markov decision processes, policy-gradient algorithms and previous probabilistic planning approaches. 1 We ...
Policies of Markov Decision Processes (MDPs) tell the next action to execute, given the current state and (possibly) the history of actions executed so far ...
In this paper we address the problem of explaining the recommendations returned by a Markov decision process (MDP) that is part of an intelligent assistant ...
We are interested in the resolution of general Markov decision pro- cesses (MDP) with factored state and action spaces, called FA-FMDP.
We present a simple approach for computing reasonable policies for factored Markov decision processes (MDPs), when the opti- mal value function can be ...
This paper proposes a novel and efficient approximate method to represent the exponentially many constraints.