YJ Choe
 |
YJ Choe |
About Me
I am a fourth-year joint Ph.D. student in Statistics and Machine Learning at Carnegie Mellon University. I work on problems in feedback-driven machine learning and active learning for deep learning, under the able supervision of Aarti Singh. I am also fortunate to have worked with Aaditya Ramdas recently on the problem of (sequentially) comparing sequential forecasters. From mid-2017 to 2020, I took a lengthy leave of absence from the program to work as a Research Scientist at Kakao Brain and Kakao. In Spring 2017, I received my M.S. in Machine Learning as part of the Ph.D. program, advised by Aarti Singh, Sivaraman Balakrishnan, and Timothy Verstynen. Before that, I got my B.S. in Mathematics and Computer Science at the University of Chicago, where I worked with John Lafferty and at the Knowledge Lab.
In terms of research, I am broadly interested in statistical machine learning and deep learning, with a recent focus on feedback-driven machine learning, sequential decision making, and forecasting. At Kakao Brain and Kakao, I worked on various projects involving deep learning and (multilingual) natural language processing. During my first two years at CMU, I worked on problems in high-dimensional statistics with applications to neuroscience, and I also dabbled with statistical analyses of neural networks. Please refer to my research page or CV for further information.
I was born and raised in Seoul, South Korea. My legal first name is Yo Joong (요중), which I prefer to abbreviate as YJ in English.
News
2021.10: Our new preprint, Comparing Sequential Forecasters, is released on arXiv! Code accompanying the paper is available here. This is joint work with Aaditya Ramdas.
2021.08: I am now a fourth-year Ph.D. student at CMU.
2021.02: I returned to Carnegie Mellon University for my third year as a Ph.D. student.
2020.11: I left Kakao Brain.
2020.10: I got married!
2020.09: Our paper, KorNLI and KorSTS: New Benchmark Datasets for Korean Natural Language Understanding, is accepted to Findings of ACL: EMNLP 2020. Dataset accompanying the paper is available here. Work done at Kakao Brain.
2020.07: I presented our work, An Empirical Study of Invariant Risk Minimization, at the ICML 2020 Workshop on Uncertainty & Robustness in Deep Learning. Code accompanying the paper is available here. Work done at Kakao Brain.