Hi. I'm Se-Young Yun.

I'm an associate professor of Graduate School of AI at KAIST and a member of OSI lab. My research interests lie in mathematical modeling and analysis on networks at large, with a specific focus on clustering and learning problems.
Learn about what I do

Contact me : yunseyoung at gmail or yunseyoung at kaist dot ac dot kr

Here's my history.


BS: Electrical Engineering, KAIST, 2002.3~2006.2
Ph.D: Electrical Engineering, KAIST, 2006.3~2012.2


Outstanding Reviewer Awards (NIPS 2016)
Best Paper Awards (ACM MobiHoc 2013)
Summa Cum Laude (KAIST 2006)

Work History

KAIST Graduate School of AI. Daejeon, South Korea (2019.5 ~)
KAIST IE Dept. Daejeon, South Korea (2017.7 ~)
Los Alamos National Laboratory, Los Alamos, NM, USA (2016.4~2017.7)
Microsoft Research, Cambridge, UK (2015.6~2016.3)
MSR-INRIA Joint Research Center, Paris, France (2014.4~2015.4)
KTH, Stockholm, Sweden (2013.2~2014.3)
KAIST, Daejeon, South Korea (2012.2~2013.2)

Here's my publications.

For full pulication lists, my google scholar page and dblp page

Meta Learning

"BOIL: Towards Representation Change for Few-shot Learning" with Jaehoon Oh, Hyungjun Yoo, and ChangHwan Kim, ICLR2021

Community Detection, Clustering

"Clustering in Block Markov Chains" with Alexandre Proutiere and Jaron Sanders, in Annals of Statistics
"Optimal Sampling and Clustering in the Stochastic Block Model" with Alexandre Proutiere, NeurIPS2019
"Optimal Cluster Recovery in the Labeled Stochastic Block Model" with Alexandre Proutiere, NIPS2016 arXiv
"Streaming, Memory Limited Algorithms for Community Detection" with Marc Lelarge and Alexandre Proutiere, NIPS 2014
"Community Detection via Random and Adaptive Sampling" with Alexandre Proutiere, COLT 2014

Low-Rank Approximation

"Low Rank Approximation for Streaming and Distributed Data" with Jaeseong Jeong and Alexandre Proutiere (submitted)
"Fast and Memory Optimal Low-Rank Matrix Approximation" with Marc Lelarge and Alexandre Proutiere, NIPS 2015


"Convergence Rates of Gradient Descent and MM Algorithms for Bradley-Terry Models" with Milan Vojnovic and Kaifang Zhou, AISTATS2020
"Parameter Estimation for Generalized Thurstone Choice Models" with Milan Vojnovic, ICML2016

Reinforcement Learning/Bandit

"Improved Regret Bounds of Bilinear Bandits using Action Space Dimension Analysis" with Jang Kyoungseok, Kwang-Sung Jun, and Wanmo Kang, ICML2021
"Minimal Regret in Online Rec- ommendation Systems" with Kaito Ariu, Narae Ryu, and Alexandre Proutiere, NeurIPS2020
"Reinforcement with fading memories" with Kuang Xu, in Mathematics of Operations Research. Preliminary version: Proceedings of ACM SIGMETRICS 2018.

Submodular Optimizatoin

"Test Score Algorithms for Budgeted Stochastic Utility Maximization" preprint
"Sketching with Test Scores and Submodular Maximization" with Shreyas Sekar and Milan Vojnovic, in Management Science arXiv

MCMC, BP, Graphical Model

"Spectral Approximate Inference" with Sejun Park, Eunho Yang, and Jinwoo Shin, ICML 2019
"Distributed Coordination Maximization over Networks: A Stochastic Approximation Approach", with Hyeryung Jang, Jinwoo Shin, and Yung Yi, MobiHoc 2016
"Distributed Medium Access over Time-varying Channels", with Jinwoo Shin and Yung Yi, IEEE/ACM ToN
"CSMA using the Bethe Approximation: Scheduling and Utility Maximization" with Jinwoo Shin and Yung Yi, IEEE ToI