Recently, after closely following the outcomes of the widely publicized AlphaGo match against Lee Sedol, I spent some of my time in the midst of my summer internship taking online classes and familiarizing myself with some of the concepts and latest research in Reinforcement Learning. There are already tons and tons of well-written, in-depth tutorials and blog posts on this field. I will be aggregating the references that I have found, which are in the form of blogposts, videos, papers, and online lecture material. Below is a list of my references.
If you haven’t already, check out the OpenAI Gym, which is a sweet leaderboard for simulators that cover various RL challenges. In addition to being a leaderboard, it also serves as forum to share instructional guides for implementing some of the popular RL algorithms.
Lastly, I also have a sweet repo of papers in my Dropbox that I’ve been reading/planning to read.
After doing some more searching, I found similar repositories to what I have constructed above (which I haven’t yet checked out entirely), that also might be useful to reference.
Finally, one really cool place where some of this research is being applied in interesting ways is Project Magenta from Google Brain, which, among other things, does music generation using LSTMs.