Add to ORKGWe present a provably near-optimal algorithm for reinforcement learn-ing in Markov decision processes in which there is a natural metric on the state space that allows the construction of accurate local models. Our algorithm is a generalization of the E3 algorithm of Kearns and Singh, and assumes a black box for approximate planning. Unlike the original E 3, our algorithm finds a near optimal policy in an amount of time that does not directly depend on the size of the state space, but instead de-pends on the covering numbers of the state space, which are informally the number of neighborhoods in the state space required for accurate lo-cal modeling at a chosen resolution. 1 Introduction, Motivatio
The impetus for exploration in reinforcement learning (RL) is decreasing uncertainty about the envir...
A key component of any reinforcement learning algorithm is the underlying representation used by the...
We study online reinforcement learning for finite-horizon deterministic control systems with arbitra...
We present metric- E3 a provably near-optimal algorithm for reinforcement learning in Markov decisio...
We present metric- � , a provably near-optimal algorithm for reinforcement learning in Markov decisi...
Address email We present an approximation scheme for solving Markov Decision Processes (MDPs) in whi...
Increasing attention has been paid to reinforcement learning algorithms in recent years, partly due ...
We present a class of metrics, defined on the state space of a finite Markov decision process (MDP)...
The problem of selecting the right state-representation in a reinforcement learning problem is consi...
In most practical applications of reinforcement learning, it is untenable to maintain direct estimat...
Recent advancements in model-based reinforcement learning have shown that the dynamics of many struc...
Recent advancements in model-based reinforcement learn-ing have shown that the dynamics of many stru...
We present new algorithms for reinforcement learning, and prove that they have polynomial bounds on ...
Reinforcement Learning (RL) in finite state and action Markov Decision Processes is studied with an ...
We address the problem of autonomously learning controllers for vision-capable mo...
The impetus for exploration in reinforcement learning (RL) is decreasing uncertainty about the envir...
A key component of any reinforcement learning algorithm is the underlying representation used by the...
We study online reinforcement learning for finite-horizon deterministic control systems with arbitra...
We present metric- E3 a provably near-optimal algorithm for reinforcement learning in Markov decisio...
We present metric- � , a provably near-optimal algorithm for reinforcement learning in Markov decisi...
Address email We present an approximation scheme for solving Markov Decision Processes (MDPs) in whi...
Increasing attention has been paid to reinforcement learning algorithms in recent years, partly due ...
We present a class of metrics, defined on the state space of a finite Markov decision process (MDP)...
The problem of selecting the right state-representation in a reinforcement learning problem is consi...
In most practical applications of reinforcement learning, it is untenable to maintain direct estimat...
Recent advancements in model-based reinforcement learning have shown that the dynamics of many struc...
Recent advancements in model-based reinforcement learn-ing have shown that the dynamics of many stru...
We present new algorithms for reinforcement learning, and prove that they have polynomial bounds on ...
Reinforcement Learning (RL) in finite state and action Markov Decision Processes is studied with an ...
We address the problem of autonomously learning controllers for vision-capable mo...
The impetus for exploration in reinforcement learning (RL) is decreasing uncertainty about the envir...
A key component of any reinforcement learning algorithm is the underlying representation used by the...
We study online reinforcement learning for finite-horizon deterministic control systems with arbitra...