Kangwook Lee

Assistant Professor
Department of Electrical and Computer Engineering
Department of Computer Sciences (by courtesy)
University of Wisconsin-Madison

Affiliations:
MLOPT Research Group
Institute for Foundations of Data Science (IFDS)

Email: +first_name.last_name+ at +wisc dot edu
Office: M1002G, Engineering Centers Building
Phone: +1-608-265-4841
Webpage, (by publication date) (by citation)

News

• (Mar. 2022) 2022 KSEA Young Investigator Grants.

New preprints

• (Apr. 2022) Breaking Fair Binary Classification with Optimal Flipping Attacks.
• (Feb. 2022) Rare Gems: Finding Lottery Tickets at Initialization.
• (Jan. 2022) Improved Input Reprogramming for GAN Conditioning.
• (Jan. 2022) GenLabel: Mixup Relabeling using Generative Models.
• (Oct. 2021) Improving Fairness via Federated Learning.

New publications

• (May 2022) One paper is accepted to ICML 2022.
  • GenLabel: Mixup Relabeling using Generative Models
• (Apr. 2022) One paper is accepted to ISIT 2022
  • Breaking Fair Binary Classification with Optimal Flipping Attacks
• (Apr. 2022) One paper is published to GetMobile: Mobile Computing and Communications
  • The Roaming Edge and its Applications
• (Mar. 2022) One paper is accepted to ACL 2022 Workshop LT-EDI
  • Debiasing Pre-Trained Language Models via Efficient Fine-tuning
• (Jan. 2022) One paper is accepted to ICLR 2022.
  • Permutation-Based SGD: Is Random Optimal?
• (Jan. 2022) Two papers are accepted to FL-AAAI-22.
  • Federated Unsupervised Clustering with Generative Models
  • Improving Fairness via Federated Learning
• (Sep. 2021) Two papers are accepted to NeurIPS 2021.
  • Sample Selection for Fair and Robust Training
  • Gradient Inversion with Generative Image Prior
• (June 2021) Two papers are accepted to ICML 2021 Workshops.
  • Gradient Inversion with Generative Image Prior
  • Empirical Study on the Effective VC Dimension of Low-rank Neural Networks
• (May 2021) Two papers are accepted to ICML 2021.
  • Coded-InvNet for Resilient Prediction Serving Systems long oral
  • Discrete-Valued Latent Preference Matrix Estimation with Graph Side Information
• (Jan. 2021) One paper is accepted to MLSys 2021.
  • Accordion: Adaptive Gradient Communication via Critical Learning Regime Identification.
• (Jan. 2021) One paper is accepted to ICLR 2021.
  • FairBatch: Batch Selection for Model Fairness

Invited talk

• 2022
  • (Apr) USC EE, “On a bilevel optimization approach to fair classification”
  • (Apr) UC Santa Barbara CCDC Seminar Series, “On a bilevel optimization approach to fair classification”
  • (Mar) KAIST EE Colloquium, “On Trustworthy and Scalable Machine Learning”
  • (Mar) Informs Optimization Soceity conference 2022, “On a bilevel optimization approach to fair classification”
  • (Feb) UC Berkekley BLISS Seminar Series, “Improving Fairness via Federated Learning”
• 2011
  • (Dec) KAIST AI International Symposium, “Improving Fairness via Federated Learning”
  • (Oct) KRAFTON Developer Connect, “On Trustworthy Machine Learning”
  • (Sep) Visiting Professor Series, American Family Insurance.
  • (Jun) AI institute of POSTECH, “Information Theory and Coding for Trustworthy and Scalable Machine Learning”
  • (Jun) The Shannon meets Turing Colloquium @ Seoul National University, “Information Theory and Coding for Trustworthy and Scalable Machine Learning”
  • (Jun) KRAFTON, Inc. and Furiosa.ai, “Recent Trends of AI Research”
  • (Apr) IFDS Ethics & Algorithms SIG, “Fairness in AI”
  • (Feb) Invited lecture on Fairness in AI as a part of Lectures on Machine Learning based ICT, Korea Information and Communications Society