Tiankai Xie
XAI | Visual Analytics | HCI
I design and develop visual analytics tools to help people understand and reason about machine learning models. My research interests include Visual Analytics, Explainable AI, and Human-Computer Interaction. My goal is to make machine learning more transparent, fair, and understandable for everyone so that we can build more trustworthy and equitable AI systems.
Education
Ph.D. in Computer Science
Aug 2018 - Aug 2023
Arizona State University, Tempe, AZ
Thesis: Explaining the Vulnerability of Machine Learning through Visual Analytics
Committee: Ross Maciejewski (Chair), Huan Liu, Chris Bryan, Hanghang Tong
M.S. in Computer Science
Aug 2015 - May 2017
Stevens Institute of Technology, Hoboken, NJ
Specialization: Distributed Systems
B.S. in Computer Science
Sep 2011 - Jun 2015
Beijing Forestry University, Beijing, China
Thesis: Plant-based Ecommerce System
Research Experience
Postdoctoral Researcher
Aug 2023 - Present
VADER Lab, Arizona State University, Tempe, AZ
Working as a postdoctoral researcher in the VADER Lab, focusing on variety research topics, including Visualization Authoring System, Topological Data Analysis, Machine Learning Security and Privacy, and Explainable AI, etc. Mentoring graduate and undergraduate students.
Visiting Researcher
May 2024 - July 2024
Lawrence Berkeley National Laboratory, Berkeley, CA
Worked with the research team of LBNL on the project of visualizing the high-dimensional functions for scientific machine learning.
Research Assistant
Aug 2018 - May 2023
VADER Lab, Arizona State University, Tempe, AZ
Work as a research associate for the VADER Lab with the research topics in Explainable AI and Visual Analytics. Current dissertation topic is Explaining the vulnerabilities of machine learning models through visual analytics. Passed dissertation prospectus in Fall 2021.
Industry Experience
Data Scientist Intern
May 2021 - Aug 2021
Epsilon Data Management, LLC., Chicago, IL
Designed and implemented the algorithm to extract highlights from the aggregated audience data across 2500+ companies. Designed, implemented and integrated the Intelligent Audience Profile (IAP) visualization view driven by the designed highlighting algorithm into the DiME visual analytics platform.
Co-founder
2017 - 2018
Robotgyms, Inc., San Mateo, CA
Developed and grew Robotgyms Inc.'s robotics education programs as a co-founder from scratch. Created a comprehensive, systematic robotics curriculum tailored for K-12 students, which included over 50 engaging hands-on projects. Organized the curriculum and established effective facility management methods to ensure smooth operations across multiple class locations. Developed online resources to support students' offline assignments and delivered weekly lectures to classes of more than 100 students. Collaborated with partners to drive sign-ups for the robotics summer camp program through SEO and SMO strategies, resulting in over 200 enrollments and contributing to a substantial 40% quarter-on-quarter revenue increase for the company.
Publications
Explaining the Vulnerability of Machine Learning through Visual Analytics
2023
Ph.D. Thesis
📄 Paper
BibTeX
@phdthesis{xie2023explaining,
title={Explaining the Vulnerabilities of Machine Learning through Visual Analytics},
author={Xie, Tiankai},
year={2023},
school={Arizona State University}
}
InfoFair: Information-Theoretic Intersectional Fairness
2022
Jian Kang, Tiankai Xie, Xintao Wu, Ross Maciejewski, Hanghang Tong
IEEE International Conference on Big Data (Big Data), 2022
📄 Paper 🔗 DOI
BibTeX
@INPROCEEDINGS{10020588,
author={Kang, Jian and Xie, Tiankai and Wu, Xintao and Maciejewski, Ross and Tong, Hanghang},
booktitle={2022 IEEE International Conference on Big Data (Big Data)},
title={InfoFair: Information-Theoretic Intersectional Fairness},
year={2022},
volume={},
number={},
pages={1455-1464},
doi={10.1109/BigData55660.2022.10020588}}
FairRankVis: A Visual Analytics Framework for Exploring Algorithmic Fairness in Graph Mining Models
2021
Tiankai Xie, Yuxin Ma, Hanghang Tong, Ross Maciejewski
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2021
📄 Paper 🎥 Talk 💻 Code 🔗 DOI
BibTeX
@article{xie2021fairrankvis,
title={Fairrankvis: Visual analytics for auditing the fairness of graph-based ranking},
author={Xie, Tiankai and Ma, Yuxin and Tong, Hanghang and Maciejewski, Ross},
journal={IEEE transactions on visualization and computer graphics},
volume={27},
number={2},
pages={1589--1599},
year={2021},
publisher={IEEE}
}
Auditing the Sensitivity of Graph-based Ranking with Visual Analytics
2020
Tiankai Xie, Yuxin Ma, Hanghang Tong, My T Thai, Ross Maciejewski
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2020
📄 Paper 🎥 Talk 💻 Code 🔗 DOI
BibTeX
@article{xie2020auditing,
title={Auditing the sensitivity of graph-based ranking with visual analytics},
author={Xie, Tiankai and Ma, Yuxin and Tong, Hanghang and Thai, My T and Maciejewski, Ross},
journal={IEEE transactions on visualization and computer graphics},
volume={26},
number={1},
pages={1075--1085},
year={2020},
publisher={IEEE}
}
Explaining Vulnerabilities to Adversarial Machine Learning through Visual Analytics
2019
Yuxin Ma, Tiankai Xie, Jundong Li and Ross Maciejewski
IEEE Transactions on Visualization and Computer Graphics (TVCG), Vancouver, Canada, 2019
📄 Paper 🎥 Talk 💻 Code 🔗 DOI
BibTeX
@article{ma2019explaining,
title={Explaining vulnerabilities to adversarial machine learning through visual analytics},
author={Ma, Yuxin and Xie, Tiankai and Li, Jundong and Maciejewski, Ross},
journal={IEEE transactions on visualization and computer graphics},
volume={26},
number={1},
pages={1075--1085},
year={2019},
publisher={IEEE}
}
Services
Reviewer
IEEE Visualization (VIS)
IEEE Transactions on Visualization and Computer Graphics (TVCG)
IEEE Computer Graphics and Applications (CGA)
IEEE Pacific Visualization Symposium (PacificVis)
Neural Information Processing Systems (NeurIPS)
References