Power of Alternating Direction Method of Multipliers(ADMM) in deep learning
Talks at Modeling and Optimization, Theory and Applications (MOPTA), Bethlehem, PA, United States
CV
Education
- Ph.D. in Computer Science, Emory University, Atlanta, Georgia, United States, 2022
- M.S. in Applied Information Technology, George Mason University, Fairfax, Virginia, United States, 2020
- B.S. in Computer Science and Technology, East China Normal University, Shanghai, China, 2012
Experience
- 2023-now: Researcher
- NEC Laboratories America
- 2017-2022: Research Assistant
- George Mason University and Emory University
Research Interests
- Deep Learning On Graph
- Graph Source Localization
- Influence Maximization
- Graph Inverse Problem
- Social Media Mining
- Vaccine Adverse Event Detection
- Petition Victory Prediction
- Nonconvex Alternating Direction Method of Multipliers(ADMM)
- Nonlinear Equality-Constrained Problems
- Multi-Convex Problems
- Deep Learning Optimization
- ADMM
- Alternating Minimization(AM)
Talks
Towards Quantized Model Parallelism for Graph-Augmented MLPs Based on Gradient-Free ADMM Framework
Talk at INFORMS Optimization Society Conference (IOS), Greenville, SC, United States
Accelerated Gradient-free Neural Network Training by Multi-convex Alternating Optimization
Talk at Association of European Operational Research Societies Conference (EURO), Espoo, Finland
Honors and Awards
- ICDM Student Travel Award, 2018, 2020
- KDD Student Travel Award, 2019
- Meritorious Winner in the Mathematical Contest in Modeling(MCM) in the United States, 2011
- Second Prize in the Contemporary Undergraduate Mathematical Contest in Modeling(CUMCM) in China, 2010
- Second Prize in the Undergraduate Mathematical Contest in Modeling by East China Normal University, 2009