I'm modeling a system whose configuration can be represented by a binary array ([1,0,0,1] or [0,1,0,0], for example), and the agent can move on a 2D space (thus having 3 DOF), and the action the agent takes depends on the position (it has to follow a predetermined trajectory). It doesn't learn, and one of my suspicions is that the configuration is normalized ([1,0,0,1] is between 0 and 1), but I can't normalize position data, because it would basically render the trajectory information useless.

Is there another way to do this, am I doing it wrong, or the problem is somewhere else?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.