I'm a fourth-year Ph.D. candidate at math department, National University of Singapore (NUS).
I'm fortunate to be advised by Professor Li Qianxiao.
Before that, I studied applied math at Fudan University.
My research interests are about sequence modelling in machine learning, language models and nonlinear systems.
Educational Background:
Ph.D. in Mathematics, National University of Singapore, Singapore, 2020-now
B.S. in Mathematics, Fudan University, Shanghai, China, 2016-2020
Published:
1. Inverse Approximation Theory for Nonlinear Recurrent Neural Networks (ICLR 2024, spotlight)
GitHub repo: Curse-of-memory
2. StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization (ICML 2024)
3. State-space models with layer-wise nonlinearity are universal approximators with exponential decaying memory (NeurIPS 2023)
GitHub repo: Awesome-state-space-models
4. Integrating Deep Learning and Synthetic Biology: A Co-Design Approach for Enhancing Gene Expression via N-terminal Coding Sequences (Nature Computational Science)
5. Efficient Hyperdimensional Computing (ECML 2023)
6. A Brief Survey on the Approximation Theory for Sequence Modelling (JML 2023)
Manuscripts:
1. LongSSM: On the Length Extension of State-space Models in Language Modelling (Submitted)
2. Improve Long-term Memory Learning Through Rescaling the Error Temporally
3. HyperSNN: A new efficient and robust deep learning model for resource constrained control applications (Submitted)
Internships:
2023.04-2023.12 Sea AI Lab Intern
2021.08-2021.10 Advance.AI R&D Intern
2019.07-2020.01 Megvii Intern
2019.01-2019.02 Goku-data Intern
Travels:
2024.05.07-2024.05.11 ICLR
2024.07.08-2024.07.12 Symposium on ML and DS
2024.07.15-2024.07.19 SciCADE
2024.07.21-2024.07.27 ICML
TA positions:
Teaching Assistant for DSA5102 (2021.08-11)
Teaching Assistant for DSA5101, DSA5102 (2020.08-11)
Skills:
Code: Python (PyTorch, JAX, TensorFlow), C/C++, Haskell
Languages:
Chinese (Mandarin) and English