Jialin Liu: Bridging Mathematics and Artificial Intelligence at the University of Central Florida
Introduction
Jialin Liu is an Assistant Professor in Statistics and Data Science at the University of Central Florida (UCF) and a member of the AI Initiative at UCF. His work focuses on the confluence of mathematics and artificial intelligence (AI), specifically the application of AI to solve computational mathematical problems. These problems include optimization, differential equations, and numerical linear algebra. While AI and data science have demonstrated considerable promise in these fields, a deep, systematic understanding of these approaches remains elusive. This necessitates the development of stable, safe, and explainable data-driven methods for mathematical applications.
Education and Background
Jialin Liu's academic journey began at Tsinghua University, where he earned his B.S. degree in Automation in 2015. He then pursued his Ph.D. in Applied Mathematics at the University of California, Los Angeles (UCLA), completing his doctorate in 2020 under the guidance of Dr. W. Yin.
Research Focus
Jialin Liu's research is centered on the convergence of mathematics and AI. He is particularly interested in applying AI to computational mathematical problems such as optimization, differential equations, and numerical linear algebra. While AI and data science have shown significant potential in these areas, a systematic and fundamental understanding of such approaches is still lacking. There is an urgent need to develop stable, safe, and explainable data-driven methods for mathematical applications. He held positions at the School of Data, Mathematical, and Statistical Sciences, UCF and Decision Intelligence Lab, Alibaba DAMO Academy (Jul. 2020 - Aug. 2021). He also worked in the Department of Mathematics, UCLA (Aug. 2015 - Jun. 2020).
The Need for Understanding
Although AI and data science have demonstrated significant potential in solving complex mathematical problems, a comprehensive and fundamental understanding of these approaches is still lacking. This gap in knowledge highlights the need for further research and development in this area.
Developing Data-Driven Methods
To address the limitations of current AI applications in mathematics, Jialin Liu emphasizes the importance of developing stable, safe, and explainable data-driven methods. These methods would ensure the reliability and transparency of AI solutions in mathematical contexts.
Key Publications
Jialin Liu's research has been published in several prestigious conferences, showcasing his contributions to the field of AI and mathematics. Some notable publications include:
- "Towards Constituting Mathematical Structures for Learning to Optimize." International Conference on Machine Learning (ICML), 2023. (With X. Chen, Z. Wang, W. Yin, and H. Cai).
- “On Representing Linear Programs by Graph Neural Networks.” International Conference on Learning Representations (ICLR), 2023. (With Z. Chen, X. Wang, J. Lu, and W. Yin).
- “On Representing Mixed-Integer Linear Programs by Graph Neural Networks.” International Conference on Learning Representations (ICLR), 2023. (With Z. Chen, X. Wang, J. Lu, and W. Yin).
- "ALISTA: Analytic Weights Are As Good As Learned Weights in LISTA." International Conference on Learning Representations (ICLR), 2019. (With X. Chen, Z. Wang, and W. Yin).
These publications demonstrate Jialin Liu's expertise in applying AI techniques, particularly graph neural networks and learned iterative shrinkage-thresholding algorithms (LISTA), to solve mathematical problems.
Focus on Optimization
Several of Jialin Liu's publications focus on the application of AI to optimization problems. Optimization is a critical area of mathematics with applications in various fields, including engineering, economics, and computer science.
Graph Neural Networks
Graph neural networks (GNNs) are a type of neural network that can operate on graph-structured data. Jialin Liu's work explores the use of GNNs to represent and solve linear and mixed-integer linear programs, which are fundamental optimization problems.
Learned Iterative Shrinkage-Thresholding Algorithm (LISTA)
The Learned Iterative Shrinkage-Thresholding Algorithm (LISTA) is a learned algorithm for solving sparse coding problems. Jialin Liu's research has shown that analytic weights can perform as well as learned weights in LISTA, simplifying the algorithm and making it more efficient.
Contributions to UCF
As an Assistant Professor at UCF, Jialin Liu contributes to the university's academic and research community through his teaching, research, and involvement in the AI Initiative. He plays a vital role in shaping the future of data science and AI at UCF. His position in the School of Data, Mathematical, and Statistical Sciences allows him to connect with students and fellow researchers, fostering collaboration and innovation.
The AI Initiative at UCF
The AI Initiative at UCF is a university-wide effort to advance AI research, education, and outreach. Jialin Liu's involvement in the AI Initiative highlights his commitment to promoting AI innovation and collaboration at UCF.
tags: #jialin #liu #university #of #central #florida
