Lee Lab @ UW Madison

Selected work/talks

[GPT/Transformers] LIFT (solving non-language ML tasks with language models): NeurIPS’22, Designing a CPU with a looped Transformer: Arxiv’23
- LIFT Github and a short summary of LIFT
- A short summary on how to design a CPU with a looped Transformer by Prof. Dimitris Papailiopoulos
- Invited talks at Samsung Advanced Institute of Technology (2022), EIRIC Korea (2022)
[Diffusion models] Finding a short-cut!: Arxiv’23, Diffusion models minimize the Wasserstein distance: NeurIPS’22
- A short summary on “diffusion models minimize the Wasserstein distance”
- Invited talks at The Coordinated Science Laboratory Student Conference (Feb. 2023), Information Theory and Applications Workshop (Feb. 2023), Information Theory and Data Science Workshop (Jan. 2023), UW Madison SILO (Jan. 2023)
[CLIP] WALIP (connecting two CLIP modles for word translation): Findings of EMNLP’22
- WALIP Github and a short summary of WALIP
[ML Fairness] FairBatch (a bilevel optimization approach to fair classification): FairBatch @ ICLR’21, Robustness + FairBatch @ NeurIPS’21, FedFB (Federated learning + FairBatch)
- Invited talks at US-Mexico Workshop on Optimization and its Applications (2023), KAIST AI International Symposium (2022), IOS’22, UCSB CCDC (2022), USC EE (2022), UC Berkeley BLISS (2021)
[Coded computation] Coded-InvNet (coded computation for deep invertible neural networks): ICML’21
- Invited talks at Seoul National University (2021), POSTECH AI (2021)

(Mar. 2023) [ICLR’23 Workshop on Mathematical and Empirical Understanding of Foundation Models] Mini-Batch Optimization of Contrastive Loss and Looped Transformers as Programmable Computers
(Feb. 2023) Prof. Lee gave/will give invited talks.
- Midwest Regional Conference (Mar. 2023)
- The Coordinated Science Laboratory Student Conference (Feb. 2023)
- Information Theory and Applications Workshop (Feb. 2023)
- Information Theory and Data Science Workshop (Jan. 2023)
- The US-Mexico Workshop on Optimization and its Applications (Jan. 2023)
(Jan. 2023) [ICLR’23] Equal Improvability: A New Fairness Notion Considering the Long-Term Impact. (Github)
(Dec. 2022) Dr. Sohn will join Yonsei University as an assistant professor! Congratulations!
(July. 2022) Prof. Lee received UW Madison ECE Grainger Faculty Scholarship Award.
(Mar. 2022) Prof. Lee received 2022 KSEA Young Investigator Grants.

Improving Fair Training under Correlation Shifts
Yuji Roh, Kangwook Lee, Steven Euijong Whang, Changho Suh
Arxiv 2023
Optimizing DDPM Sampling with Shortcut Fine-Tuning
Ying Fan, Kangwook Lee
Arxiv 2023
Looped Transformers as Programmable Computers
Angeliki Giannou, Shashank Rajput, Jy-yong Sohn, Kangwook Lee, Jason D. Lee, Dimitris Papailiopoulos
Arxiv, 2023
A Better Way to Decay: Proximal Gradient Training Algorithms for Neural Nets
Liu Yang, Jifan Zhang, Joseph Shenouda, Dimitris Papailiopoulos, Kangwook Lee, and Robert D. Nowak
Arxiv, NeurIPS’22 OPT Workshop, 2022
Outlier-Robust Group Inference via Gradient Space Clustering
Yuchen Zeng, Kristjan Greenewald, Kangwook Lee, Justin Solomon, Mikhail Yurochkin
Arxiv, 2022