Tianjiao Ding

I am a PhD student at University of Pennsylvania, advised by René Vidal. I work closely with Benjamin D. Haeffele and Yi Ma. Prior to my PhD, I spent two years as a research assistant at ShanghaiTech University, advised by Manolis C. Tsakiris and collaborating with Laurent Kneip. I received a master’s degree in applied mathematics and statistics from Johns Hopkins University, and a bachelor’s in computer science with honor from ShanghaiTech.

My research interests lie in the theoretical foundations of machine learning and data science as well as emerging applications. As such, I develop both rigorous mathematics and practical implementations in my work. In particular, I study manifold learning and clustering, 3D vision and robotics.

I love to chat about research, life, and fun. If you are a student interested in doing research with me, please email me with your CV and transcript. For master’s students and undergraduates, the recommended time investment is at least 15 hours per week.

Updates

Mar 2024	One paper accepted to ICLR ‘24 (Vienna)!
Jan 2024	Excited to give an oral talk of a paper accepted to CPAL ‘24 (Hong Kong)!
Dec 2023	I received a master’s degree in applied mathematics and statistics from Johns Hopkins University! Delighted to move forward with roots in both mathematics and computer science.

Papers

PaCE: Parsimonious Concept Engineering for Large Language Models

Jinqi Luo*, TD*, Kwan Ho Ryan Chan, Darshan Thaker, Aditya Chattopadhyay, Chris Callison-Burch, and René Vidal

In arXiv preprint arXiv:2406.04331, 2024

Media Paper Code Cite

Penn AI

@inproceedings{luo2024pace,
  title = {PaCE: Parsimonious Concept Engineering for Large Language Models},
  author = {Luo*, Jinqi and Ding*, Tianjiao and Chan, Kwan Ho Ryan and Thaker, Darshan and Chattopadhyay, Aditya and Callison-Burch, Chris and Vidal, Ren{\'e}},
  booktitle = {arXiv preprint arXiv:2406.04331},
  year = {2024},
}

Image Clustering via the Principle of Rate Reduction in the Age of Pretrained Models

Tianzhe Chu*, Shengbang Tong*, TD*, Xili Dai, Benjamin D. Haeffele, René Vidal, and Yi Ma

In International Conference on Learning Representations, 2024

Paper Code Cite

@inproceedings{chu2024image,
  title = {Image Clustering via the Principle of Rate Reduction in the Age of Pretrained Models},
  author = {Chu*, Tianzhe and Tong*, Shengbang and Ding*, Tianjiao and Dai, Xili and Haeffele, Benjamin D. and Vidal, Ren{\'e} and Ma, Yi},
  booktitle = {International Conference on Learning Representations},
  year = {2024},
}

HARD: Hyperplane ARrangement Descent

TD*, Liangzu Peng*, and René Vidal

In Conference on Parsimony and Learning, 2024

Oral Paper Code Cite

Oral presentation

@inproceedings{ding2023hard,
  title = {HARD: Hyperplane ARrangement Descent},
  author = {Ding*, Tianjiao and Peng*, Liangzu and Vidal, Ren{\'e}},
  booktitle = {Conference on Parsimony and Learning},
  year = {2024},
}

Unsupervised Manifold Linearizing and Clustering

TD, Shengbang Tong, Kwan Ho Ryan Chan, Xili Dai, Yi Ma, and Benjamin D. Haeffele

In International Conference on Computer Vision, 2023

Paper Cite

@inproceedings{ding2023unsupervised,
  title = {Unsupervised Manifold Linearizing and Clustering},
  author = {Ding, Tianjiao and Tong, Shengbang and Chan, Kwan Ho Ryan and Dai, Xili and Ma, Yi and Haeffele, Benjamin D.},
  booktitle = {International Conference on Computer Vision},
  year = {2023},
}

Understanding Doubly Stochastic Clustering

TD, Derek Lim, René Vidal, and Benjamin D. Haeffele

In International Conference on Machine Learning, 2022

Paper Cite

@inproceedings{ding2022understanding,
  title = {Understanding Doubly Stochastic Clustering},
  author = {Ding, Tianjiao and Lim, Derek and Vidal, Ren{\'e} and Haeffele, Benjamin D.},
  booktitle = {International Conference on Machine Learning},
  year = {2022},
}

Efficient Maximal Coding Rate Reduction by Variational Forms

Christina Baek, Ziyang Wu, Kwan Ho Ryan Chan, TD, Yi Ma, and Benjamin D. Haeffele

In IEEE Conference on Computer Vision and Pattern Recognition, 2022

Paper Cite

@inproceedings{baek2022efficient,
  title = {Efficient Maximal Coding Rate Reduction by Variational Forms},
  author = {Baek, Christina and Wu, Ziyang and Chan, Kwan Ho Ryan and Ding, Tianjiao and Ma, Yi and Haeffele, Benjamin D.},
  booktitle = {IEEE Conference on Computer Vision and Pattern Recognition},
  year = {2022}
}

Robust Homography Estimation via Dual Principal Component Pursuit

TD, Yunchen Yang, Zhihui Zhu, Daniel P. Robinson, René Vidal, Laurent Kneip, and Manolis C. Tsakiris

In IEEE Conference on Computer Vision and Pattern Recognition, 2020

Paper Code Cite

@inproceedings{ding2020robust,
  title = {Robust Homography Estimation via Dual Principal Component Pursuit},
  author = {Ding, Tianjiao and Yang, Yunchen and Zhu, Zhihui and Robinson, Daniel P. and Vidal, Ren{\'e} and Kneip, Laurent and Tsakiris, Manolis C.},
  booktitle = {IEEE Conference on Computer Vision and Pattern Recognition},
  pages = {6080--6089},
  year = {2020},
}

Noisy Dual Principal Component Pursuit

Tianyu Ding, Zhihui Zhu, TD, Yunchen Yang, Daniel P. Robinson, Manolis C. Tsakiris, and René Vidal

In International Conference on Machine Learning, 2019

Oral Paper Code Cite

Oral presentation

@inproceedings{ding2019noisy,
  title = {Noisy Dual Principal Component Pursuit},
  author = {Ding, Tianyu and Zhu, Zhihui and Ding, Tianjiao and Yang, Yunchen and Robinson, Daniel P. and Tsakiris, Manolis C. and Vidal, Ren{\'e}},
  booktitle = {International Conference on Machine Learning},
  pages = {1617--1625},
  year = {2019},
}

Learning to Parse Wireframes in Images of Man-Made Environments

Kun Huang, Yifan Wang, Zihan Zhou, TD, Shenghua Gao, and Yi Ma

In IEEE Conference on Computer Vision and Pattern Recognition, 2018

Paper Code Cite

@inproceedings{huang2018learning,
  title = {Learning to Parse Wireframes in Images of Man-Made Environments},
  author = {Huang, Kun and Wang, Yifan and Zhou, Zihan and Ding, Tianjiao and Gao, Shenghua and Ma, Yi},
  booktitle = {IEEE Conference on Computer Vision and Pattern Recognition},
  pages = {626--635},
  year = {2018},
}