Deep Reinforcement Learning Doesn't Work Yet

That’s why I like more “tear down” projects and papers (that kind of paper helps to open our eyes in the right criticism) than “look-my-shiny-not-reproducible-paper-project” . By Sorta Insightful