The Canonical Success Examples of Arel Are in Computer Games

The impact of our action is not always obvious, and there's hidden variables involved. The success cases of r l are in simple kind of causal examples where everything is very much visible and the causal effects are quite simple. When you start porting reinforcement learning into real life scenarios, this hundred % correspondence between action and effect doesn't translate across. So if you want to traina trading out brethren using reinforcement learning, then you all of a sudden you have a whole slegh ofer variables which you never get mel to observe.