Robot Brains Interviews - Robotic Brains

Reinforcement learning, i think, is a really interesting area. People like to you can easily conceptualize how animals or people learn that way. And so for i'm interested in how people are going to interact with robots that are learning. I was reading one of your papers on reinforcement learning with inaccurate feedback,. where robot learns from its trial and error from feed back. An see a bit more about wy tat matters and compex thereYour student Taylor kelsar fuckner has been studying the problem of what do we even do when you have an inner accurate person? What are some of the reasons that people might give inaccurate beat back?...