132
click to vote
ICANN
2009
Springer
15 years 7 months ago
2009
Springer
In a typical reinforcement learning (RL) setting details of the environment are not given explicitly but have to be estimated from observations. Most RL approaches only optimize th...