Thanks for the perspective. I have been asking this question for a few years now, and I think Deepmind's marketing has been pretty deceptive on this. What examp

Thanks for the perspective. I have been asking this question for a few years now, and I think Deepmind's marketing has been pretty deceptive on this. What examples do you have in mind where the research from playing games with RL has carried over into solving a real-world problem? Probably the state of the art has advanced as I have focused in applying supervised learning on some applied problems that interest me.