Biography
I am a postdoctoral fellow at Shanghai Jiao Tong University, under the supervision of Prof. Quanshi Zhang. I received my PH.D. degree from Sun Yat-Sen University.
My research interest mainly focuses on explainable AI, which involves two perspectives: (1) Semantic explanation, to make human understand why a DNN makes a certain prediction; (2) Mathematical explanation, to investigate and explain the representation capacity of DNN. Our ultimate goal is to bulid an explainable, safe, and controllable AI.
Download my resumé.
Interests
- Explainable AI
- Adversarial Robustness
- Causal Learning
Education
-
PhD in Applied Mathematics, 2015-2021
Sun Yat-Sen University
-
BSc in Applied Mathematics, 2021-2015
Central China Normal University
Research Experience
- Shanghai Jiao Tong University, Postdoctoral Fellow.
- Advisor: Prof. Quanshi Zhang.
- Sep 2021 - present
- Texas A & M University, Visiting student.
- Advisor: Prof. Xia Hu.
- Jan 2020 - Jan 2021
- HongKong Bapatist University, Research Assistant
- Advisor: Prof. PongChi Yuen.
- Oct 2016 - Jun 2017, and Jan 2018 - Jan 2019
Selected Publications
Quickly discover relevant content by filtering publications.
(2021).
Discovering and Explaining the Representation Bottleneck of DNNs.
arxiv 2021.
(2021).
A General Taylor Framework for Unifying and Revisiting Attribution Methods.
arxiv 2021.
(2021).
Mutual Information Preserving Back-propagation Learn to Invert for Faithful Attribution.
KDD 2021.
(2021).
A Unified Taylor Framework for Revisiting Attribution Methods.
AAAI 2021.
Academic Talks
- [2021] IJCAI 2021 Tutorial on Interpretability
- [2021] The Eighth Issue of “Artificial Intelligence Security and Privacy” Series Forum.
- [2021] The nineteenth Annual Meeting of the Chinese Society of Industrial and Applied Mathematics.