Eliezer, do you think that this search for magic and mystery always outside the known world is some side-effect of human reinforcement learning?
As far as I know it could be a loose knob on the tuning device for the exploit/explore tradeoff.
Eliezer, do you think that this search for magic and mystery always outside the known world is some side-effect of human reinforcement learning?
As far as I know it could be a loose knob on the tuning device for the exploit/explore tradeoff.