ACL
2009
13 years 9 months ago
2009
In this paper, we present a reinforcement learning approach for mapping natural language instructions to sequences of executable actions. We assume access to a reward function tha...
NIPS
2004
14 years 25 days ago
2004
We consider an MDP setting in which the reward function is allowed to change during each time step of play (possibly in an adversarial manner), yet the dynamics remain fixed. Simi...
NIPS
2007
14 years 26 days ago
2007
We study the problem of an apprentice learning to behave in an environment with an unknown reward function by observing the behavior of an expert. We follow on the work of Abbeel ...
ESANN
2008
14 years 27 days ago
2008
In this paper the application of reinforcement learning to Tetris is investigated, particulary the idea of temporal difference learning is applied to estimate the state value funct...
AAAI
2010
14 years 27 days ago
2010
The precise specification of reward functions for Markov decision processes (MDPs) is often extremely difficult, motivating research into both reward elicitation and the robust so...
AAAI
2007
14 years 1 months ago
2007
In designing Markov Decision Processes (MDP), one must define the world, its dynamics, a set of actions, and a reward function. MDPs are often applied in situations where there i...
SIGGRAPH
2010
ACM
14 years 3 months ago
2010
ACM
We present a method for inferring the behavior styles of character controllers from a small set of examples. We show that a rich set of behavior variations can be captured by dete...
ISCAS
2006
IEEE
14 years 5 months ago
2006
IEEE
— This paper describes a general approach for the unsupervised learning of behaviors in a behavior-based robot. The key idea is to formalize a behavior produced by a Motor Map dr...
PKDD
2009
Springer
14 years 6 months ago
2009
Springer
Abstract. Inverse reinforcement learning addresses the general problem of recovering a reward function from samples of a policy provided by an expert/demonstrator. In this paper, w...
ICML
2003
IEEE
15 years 7 days ago
2003
IEEE
The paper explores a very simple agent design method called Q-decomposition, wherein a complex agent is built from simpler subagents. Each subagent has its own reward function and...