Hi, I'm an Assistant Professor of Computer Science in the School of Electrical Engineering and Computer Science at Oregon State University.
The goal of my research is to make AI more interpretable and accessible, or broadly human-centered. Specifically, my research focuses on building novel interactive, data visualization tools for people to easily explore, interpret, and interact with machine learning (ML) and large datasets. To do this, my work combines data-driven scalable techniques and human-centered interactive approaches, by utilizing methods from multiple areas, including data visualization, interpretable ML, human-computer interaction (HCI), and databases.
I received a PhD in computer science at Georgia Tech. I was fortunate to work with amazing people at Google Brain and Facebook Research, resulting in deployed systems (e.g., by Facebook on deep learning visualization) and open-sourced tools (e.g., with Google Brain on deep learning education). I was supported by NSF, Google, and Facebook.
I'm looking for current OSU grad and undergrad students interested in explainable AI, data visualization, or human-AI interaction. Please email me with your CV; a brief description of your interests and experience; and why you want to work with me.
- May 2020 I'll be teaching a new grad-level course on "Data Visualization for ML". OSU students are welcome to take it!
- Feb 2020 Our Chronodes paper nominated for the 2018 ACM TiiS Best Paper.
- Feb 2020 New paper, CNN 101, accepted for CHI 2020 work-in-progress.
- Oct 2019 Presented on the evaluation of GAN Lab at EVIVA-ML Workshop at VIS.
- July 2019 New paper, FairVis, on visual analysis of ML bias accepted to IEEE VIS'19.
Ph.D. in Computer Science,
Georgia Institute of Technology, USA
Thesis: Human-Centered AI through Scalable Visual Data Analytics
Committee: Polo Chau, Sham Navathe, Alex Endert, Martin Wattenberg, Fernanda Viégas
M.S. in Computer Science and Engineering,
Seoul National University, South Korea
Thesis: Context-Aware Recommendation using Learning-to-Rank
- B.S. in Electrical and Computer Engineering, Seoul National University, South Korea 2005-2009
Industry Research Experience
- Google, Software Engineering Intern at Google Brain's People+AI Research Group Summer 2017
- Facebook, Research Intern at Applied ML Research Group Summer 2016
- Facebook, Research Intern at Applied ML Research Group Summer 2015
- ACM Trans. Interactive Intelligent Systems (TiiS) 2018 Best Paper, Honorable Mention 2020
- Google PhD Fellowship, Google AI 2018-2019
- Graduate TA of the Year in School of Computer Science, Georgia Tech Apr 2018
- NSF Graduate Research Fellowship, National Science Foundation 2014-2017
- Best Paper Award, PhD Workshop at CIKM Oct 2011
- National Scholarship for Science and Engineering, Korea Student Aid Foundation 2005-2009
CS 565. Human-Computer Interaction
This course will provide students with basic principles of and research methods in Human-Computer Interaction (HCI). Students will learn how to design and prototype user interfaces and interactive systems, based on the needs of users, and how to evaluate such interfaces and systems rigorously.
CS 539. ST/ AI: Data Visualization for ML
This course will introduce advanced state-of-the-art research on interactive data visualization for machine learning. Data visualization is a powerful way to understand and analyze how machine learning works. Students will learn how to design and develop interactive data visualization methods and tools that are interpretable to complex ML models (e.g., deep learning models), scalable to large data, and usable to a variety of users (e.g., ML researchers, practitioners like ML engineers and data scientists, non-expert learners). We will also discuss how data visualization can help address several critical human-side issues in AI (e.g., explainability, transparency, fairness, accessibility to non-experts).
Prerequisite: Any ML course (e.g., CS 534) or HCI/visualization course (e.g., CS 565, CS 458)
Publications (Latest & Greatest) Full list (h-index: 15)
CNN 101: Interactive Visual Learning for Convolutional Neural Networks
ACM SIGCHI Conference on Human Factors in Computing Systems (CHI'20 Work in Progress), 2020.
FairVis: Visual Analytics for Discovering Intersectional Bias in Machine Learning
IEEE Conference on Visual Analytics Science and Technology (VAST'19), 2019.
DOI PDF arXiv Blog Code
How Does Visualization Help People Learn Deep Learning? Evaluation of GAN Lab
Workshop on Evaluation of Interactive Visual Machine Learning Systems (EVIVA-ML at IEEE VIS'19), 2019.
GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation
IEEE Transactions on Visualization and Computer Graphics, 25(1) (VAST'18), 2019.
Open sourced with Google AI
DOI PDF Slides Code Website
Visual Analytics in Deep Learning: An Interrogative Survey for the Next Frontiers
IEEE Transactions on Visualization and Computer Graphics, 25(8), 2019.
DOI PDF Website Medium
ActiVis: Visual Exploration of Industry-Scale Deep Neural Network Models
IEEE Transactions on Visualization and Computer Graphics, 24(1) (VAST'17), 2018.
Deployed on Facebook ML Platform; Invited to present at SIGGRAPH'18 as a top VIS paper (4 total)
DOI PDF Video Slides Website
Chronodes: Interactive Multifocus Exploration of Event Sequences
ACM Transactions on Interactive Intelligent Systems (TiiS), 8(1), 2018.
Best Paper, Honorable Mention
Interactive Browsing and Navigation in Relational Databases
Proceedings of the VLDB Endowment, 9(12) (VLDB'16), 2016.
DOI PDF Slides
Visual Exploration of Machine Learning Results using Data Cube Analysis
Workshop on Human-In-the-Loop Data Analytics (HILDA at SIGMOD'16), 2016.
Deployed on Facebook ML Platform
DOI PDF Slides