Gwonsoo Che

On Leave

I am currently inactive in terms of academic research.

For more information, check out my LinkedIn profile.

Me

I am a Ph.D. student at School of Computing, KAIST. My advisor is Hongseok Yang.

Research

I work on learning and inference problems related to probabilistic programming. My current research focus is to answer the question about how we can make an inference algorithm learn from experience, and generalize across probabilistic programs.

Outside my main research, I like reading books and watching videos about math, neuroscience, and general science. Such activities are partially related to my interest in a better scientific understanding of intelligence.

Papers

  1. Che, G. and Yang, H. Meta-Learning an Inference Algorithm for Probabilistic Programs. NeurIPS 2021 AIPLANS Workshop [paper]

  2. Lim, H., Che, G., Lee, W., and Yang, H. Differentiable Algorithm for Marginalising Changepoints. AAAI 2020. [paper]

  3. Chae, K., Oh, H., Heo, K., and Yang, H. Automatically Generating Features for Learning Program Analysis Heuristics for C-Like Programs. OOPSLA 2017. [paper]

Contact

Email: gwonsoo <dot> che <at> gmail <dot> com

Mail: KAIST School of Computing (E3-1), 291 Daehak-ro, Yuseong-gu, Daejeon, Korea 34141

Others: [LinkedIn]