This blog started out as a means to both keep track of papers I have read as well as communicating research ideas with my advisor, Michael Littman, as well as my lab, RL3. Over time, it has transitioned to focus more heavily on the former, because the 1-sided discussion of a blog (even with comments) doesn’t seem to be well suited when real discussions are needed.
It is great to see people from all over the world reading content, as I never expected it to be of use for more than a tiny handful of people.
I am a PhD candidate at Rutgers University, where I study reinforcement learning with Michael Littman in the RL3 lab. My primary interests are planning in large spaces, especially continuous domains, which I feel are particularly underrepresented in the RL literature. I am particularly motivated by robust, simple methods that solve hard problems.