I feel that on some level, human knowledge/interaction can be similarly reduced to a basic mechanistic nature. Our prior experiences play the part of training d

I feel that on some level, human knowledge/interaction can be similarly reduced to a basic mechanistic nature. Our prior experiences play the part of training data, and our responses can be reduced to mechanistic processing of stimuli. I believe integrating reinforcement learning would address the last point. I have to say though, humans are much better at trolling.