I am currently a PostDoc at University of Wisconsin-Madison, working with Prof. Robert Nowak. Prior to joining UW-Madison, I received my Ph.D. in ECE department at UT Austin, where I had wonderful years of learning and working with my advisor Prof. Constantine Caramanis. I got my Bachelor’s Degree in Electrical and Computer Engineering from Seoul National University (SNU) in 2016.
I am broadly interested in theoretical aspects of machine/statistical learning and optimization, with emphasis on fundamental questions well motivated by real applications. Specifically, I have been focused on theoretical questions arising from statistical inference of latent variable models, and reinforcement learning in partially observable systems. I believe in some nice connections between these two problems. I am also interested in other research topics such as robust statistics, multitask/meta learning, and stochastic approximation.
(7/15/22) I gave a talk in a Beyond RL Workshop at TTIC.
(7/12/22) I successfully defended my thesis!
(5/15/22) One paper accepted to ICML 2022.
(4/15/22) Presented my work on Reward-Mixing MDPs at MLOPT Idea Seminar in Wisconsin-Madison.
(9/28/21) Two papers accepted to NeurIPS 2021!
(5/18/21) Presented my work on Latent MDPs at virtual RL Theory Seminar.
(3/3/21) I received a UT graduate continuing fellowship for the 2021-2022 academic year.
(1/15/21) Done with my proposal, and now become a Ph.D. candidate.
Other aspects of life
I used to play the piano, mostly the pieces of Beethoven and Chopin. My favoriate song to play is Chopin’s Etude 25-11. I am also a huge fan of rock and metal musics. My favorite bands are Skid row (in late 80s-90s), Slipknot, Stryper (glad that they are still active!). These days, I often listen to Game OST when I study or relax.
I am a huge fan of the game of Go, and play it online in my free time. I became interested in playing Go since I watched the super match between Lee Sedol and Alphago, just as much as that event interested me to pursue my Ph.D. in machine learning, especially for RL.