ICANN
2009
Springer
13 years 10 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...