Main  Blog  CV  Projects  Publications 

Here you will find writeups for some of my personal projects, including descriptions of physical, software, and/or abstract objects that I have created. See also my Github

An implementation of the board game Welcome to the Dungeon with reinforcement learning agents.

A crowdsourced list of details for "Sherlock Holmes"-style "deduction".

Strictly speaking, this part of the inference process is "abduction" not deduction. In the spirit of Sutton's bitter lesson, perhaps attaining Sherlock level insight depends on collecting a superhuman level of life experience as opposed to being primarily bottlenecked by fluid intelligence. Since it is not practical to live many lives in parallel, why not aggrate experience from a large group of people? By studying this centralized resource, "anyone" can become more Sherlockian! If you're interested, give it a read and contribute some wisdom!