Zhe Tao 陶浙

Ph.D. Candidate
Davis Automated Reasoning Group
Department of Computer Science
University of California, Davis

Email: zhetao#ucdavis.edu | zhe#zhe-tao.com


About Me

I’m a Ph.D. candidate in the Davis Automated Reasoning Group at UC Davis, advised by Prof. Aditya V. Thakur. I’m interested in programming languages, automated reasoning and machine learning.


Recent News


Recent Publications

  1. Provable Editing of Deep Neural Networks using Parametric Linear Relaxation
    Zhe Tao and Aditya V. Thakur
    Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems (NeurIPS 2024). December 2024.

  2. Towards Verifying Exact Conditions for Implementations of Density Functional Approximations
    Sameerah Helal, Zhe Tao, Cindy Rubio-González, Francois Gygi and Aditya V. Thakur
    SC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis

  3. Provable Repair of Vision Transformers [ Paper | BibTeX | GitHub ]
    Stephanie Nawas, Zhe Tao and Aditya V. Thakur
    The 7th International Symposium on AI Verification (SAIV). 2024.

  4. Architecture-Preserving Provable Repair of Deep Neural Networks. [ Paper | BibTeX | GitHub | arXiv ]
    Zhe Tao, Stephanie Nawas, Jacqueline Mitchell and Aditya V. Thakur
    44th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2023). June 2023.

  5. SyReNN: A Tool for Analyzing Deep Neural Networks. [ Paper | BibTeX | GitHub ]
    Matthew Sotoudeh, Zhe Tao and Aditya V. Thakur
    International Journal on Software Tools for Technology Transfer (STTT).

  6. DICE*: A Formally Verified Implementation of DICE Measured Boot. [ Paper | BibTeX | GitHub ]
    Zhe Tao, Aseem Rastogi, Naman Gupta, Kapil Vaswani and Aditya V. Thakur
    30th USENIX Security Symposium (USENIX Security ’21). August 2021.

Equal contribution.


Talks

  1. Algorithms and Applications for Provable Repair of Deep Neural Networks. (Co-host) [ Website ]
    45th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2024). June 2024. Copenhagen, Denmark.

  2. Architecture-Preserving Provable Repair of Deep Neural Networks. [ Website ]
    44th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2023). June 2023. Orlando, Florida, United States.

  3. DICE*: A Formally Verified Implementation of DICE Measured Boot. [ website ]
    30th USENIX Security Symposium (USENIX Security ’21). August 2021. (Virtual).


Software


Education


Teaching Experience


Services