Seiyun Shin

Research Interests

I seek to gain insight into fundamental problems that are practically relevant. My research interests lie at the intersection of theoretical machine learning, algorithm design, and information theory. Within these disciplines, I seek to achieve two closely-related goals. One is to characterize the fundamental limits of the amount and quality of data required for reliable estimation and learning; the other is to develop computationally efficient algorithms that can provably achieve these limits.. Specifically, I am working on establishing sample complexity in GraphML (e.g., graph neural networks), instance-adaptive algorithms, and multi-armed bandits.

Conference Papers

• S. Shin, I. Shomorony, and Han Zhao, "Efficient learning of linear graph neural networks via node subsampling," NeurIPS 2023 (To appear).

• S. Shin, H. Zhao, and I. Shomorony, "Adaptive power method: Eigenvector estimation from sampled data," ALT 2023 (slides).

• H. Park*, S. Shin*, KS. Jun, and J. Ok, "Transfer learning in bandits with latent continuity," ISIT 2021 (*equal contributions).

• S. Shin, R. Heckel, and I. Shomorony, "Capacity of the erasure shuffling channel," ICASSP 2020.

• S. Shin and C. Suh, "Capacity of a two-way function multicast channel,'' Allerton Conference 2017.

• K. Min, M. Jung, S. Shin, S. Kim, and S. Choi, "System level simulation of mmWave based mobile Xhaul networks," IEEE VTC Spring 2017.

• S. Shin and C. Suh, "Two-way function computation,'' Allerton Conference 2014.

Journal Papers

• H. Park, S. Shin, KS. Jun, and J. Ok, "Transfer learning in bandits with latent continuity," submitted to the IEEE Transactions on Information Theory, 2021.

• S. Shin and C. Suh, "Two-way function computation," IEEE Transactions on Information Theory, Vol 66, No 2, 2020.

Honors and Awards

• Awarded Mavis Future Faculty Fellow 2023–2024

• Nominated Olesen Award for Excellence in Undergraduate Teaching 2022

• Awarded Kwanjeong Educational Foundation Fellowship 2019–Present

• Awarded Academic Excellence Scholarship from KAIST EE Department 2009–2011

• Awarded Korea Government Fellowship 2008–2015

Academic Service

• Reviewer for Conference on Neural Information Processing Systems (NeurIPS)

• Reviewer for International Conference on Machine Learning (ICML)

• Reviewer for International Conference on Learning Representations (ICLR)

• Reviewer for IEEE Transactions on Information Theory

• Reviewer for IEEE International Symposium on Information Theory (ISIT)

• Reviewer for IEEE Information Theory Workshop (ITW)

• Reviewer for International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

• Reviewer for IEEE Wireless Communications and Networking Conference (WCNC)

• General Chair for the Coordinated Science Laboratory (CSL) Student Conference 2023 in UIUC

• Machine Learning Session Chair for the CSL Student Conference 2023

Teaching

I have taught the following courses as a teaching assistant (TA) in my undergraduate and graduate years:

@ UIUC:

• ECE313: Probability with Engineering Applications (Spring 2022)
  nomination for Olesen Award for excellence in undergraduate teaching

• ECE563: Information Theory (Fall 2021)

@ KAIST:

• EE321: Communication Engineering (Spring 2014)

• EE623: Information Theory (Fall 2013)

• EE202: Signals and Systems (Fall 2011)