Add to ORKGWe present metric- � , a provably near-optimal algorithm for reinforcement learning in Markov decision processes in which there is a natural metric on the state space that allows the construction of accurate local models. The algorithm is a generalization of the � algorithm of Kearns and Singh, and assumes a black box for approximate planning. Unlike the original � , metric-� finds a near optimal policy in an amount of time that does not directly depend on the size of the state space, but instead depends on the covering number of the state space. Informally, the covering number is the number of neighborhoods required for accurate local modeling. 1 Introduction, Motivatio
The impetus for exploration in reinforcement learning (RL) is decreasing uncertainty about the envir...
Planning plays an important role in the broad class of decision theory. Planning has drawn much atte...
Recent advancements in model-based reinforcement learning have shown that the dynamics of many struc...
We present a provably near-optimal algorithm for reinforcement learn-ing in Markov decision processe...
We present metric- E3 a provably near-optimal algorithm for reinforcement learning in Markov decisio...
Address email We present an approximation scheme for solving Markov Decision Processes (MDPs) in whi...
In most practical applications of reinforcement learning, it is untenable to maintain direct estimat...
We present a class of metrics, defined on the state space of a finite Markov decision process (MDP)...
A key component of any reinforcement learning algorithm is the underlying representation used by the...
Reinforcement Learning (RL) in finite state and action Markov Decision Processes is studied with an ...
Increasing attention has been paid to reinforcement learning algorithms in recent years, partly due ...
We address the problem of autonomously learning controllers for vision-capable mo...
We study online reinforcement learning for finite-horizon deterministic control systems with arbitra...
We address the problem of autonomously learning controllers for vision-capable mobile robots. We ext...
We introduce a model-free algorithm for learning in Markov decision processes with parameterized act...
The impetus for exploration in reinforcement learning (RL) is decreasing uncertainty about the envir...
Planning plays an important role in the broad class of decision theory. Planning has drawn much atte...
Recent advancements in model-based reinforcement learning have shown that the dynamics of many struc...
We present a provably near-optimal algorithm for reinforcement learn-ing in Markov decision processe...
We present metric- E3 a provably near-optimal algorithm for reinforcement learning in Markov decisio...
Address email We present an approximation scheme for solving Markov Decision Processes (MDPs) in whi...
In most practical applications of reinforcement learning, it is untenable to maintain direct estimat...
We present a class of metrics, defined on the state space of a finite Markov decision process (MDP)...
A key component of any reinforcement learning algorithm is the underlying representation used by the...
Reinforcement Learning (RL) in finite state and action Markov Decision Processes is studied with an ...
Increasing attention has been paid to reinforcement learning algorithms in recent years, partly due ...
We address the problem of autonomously learning controllers for vision-capable mo...
We study online reinforcement learning for finite-horizon deterministic control systems with arbitra...
We address the problem of autonomously learning controllers for vision-capable mobile robots. We ext...
We introduce a model-free algorithm for learning in Markov decision processes with parameterized act...
The impetus for exploration in reinforcement learning (RL) is decreasing uncertainty about the envir...
Planning plays an important role in the broad class of decision theory. Planning has drawn much atte...
Recent advancements in model-based reinforcement learning have shown that the dynamics of many struc...