Xin Yu Tan
I'm a 3rd year undergraduate at UC Berkeley studying Electrical Engineering and Computer Science.

I'm interested in artificial intelligence and computer vision; brain machine interfaces and virtual reality; and UI/UX in mobile, web, and game development.

Currently I'm doing research in robotics and deep reinforcement learning at UC Berkeley's Robotics Learning Lab. I work with PhD student Chelsea Finn and under the supervision of Professor Pieter Abbeel. Below are links to our ICRA 2016 paper on deep spatial autoencoders:

Deep Spatial Autoencoders for Visuomotor Learning
Chelsea Finn, Xin Yu Tan, Yan Duan, Trevor Darrell, Sergey Levine, Pieter Abbeel
(Previous title: "Learning Visual Feature Spaces for Robotic Manipulation with Deep Spatial Autoencoders")

I've also developed the user interface for the open source Guided Policy Search repository, which is used by numerous researchers in the RLL and other labs.

I've contributed to multiple programming projects. Currently, I am working on a 2D real-time strategy Android game called League of Legions, where different colored pixel armies battle for dominance. As part of a class project, I created Gitlet, which is a miniature version of Git that runs in the command line. I was the lead programmer for Cyan Knight, a 2D platformer game developed using Unity 3D game engine. I've also worked on a personal project called Chess 1.0, which used Java and Swing to develop a two-player chess program compliant with all the rules of chess.

I'm also active in community service, especially in the education domain. I was the president of ANova, a student organization dedicated to improving computer science education in low-income communities across the Bay Area. I also was the webmaster for and active volunteer in Cal Rotaract, a community service organization that hosts weekly events such as restoring natural areas, helping the homeless, and much more. I used to be a lab assistant for the data structures class CS 61B.

Contact:
Email: xinyutan@berkeley.edu

Links:
Resume: [embedded view] [pdf view]