Daiwei Chen (陈戴纬)

ECE PhD @ UW-Madison, focus on AI research.

profile.png

     

I'm a first year Electrical and Computer Engineering (ECE) PhD in Machine Learning and Optimization Theory (MLOPT) Research Group at the University of Wisconsin-Madison.

Under the guidance of my advisor, Prof. Ramya Korlakai Vinayak, my research focuses on AI alignment, Foundation Models, Representation Learning, and Preference Learning.

Before starting my PhD journey, I obtained my Master's degree in Systems Engineering from the University of Pennsylvania under the guidance of Prof. Pratik Chaudhari, and received my Bachelor of Science in Psychology from Zhejiang University.

news

Jun 19, 2024 Our paper Pluralistic Alignment Framework for Learning from Heterogeneous Preferences has been accepted by 2O24 ICML TF2M Workshop and MHFAIA Workshop!
Jun 14, 2024 Our recent work on diverse/pluralistic alignment is currently under review. If you are interested in this topic, welcome to check out the version available on arXiv!
Feb 16, 2024 Our blogpost Unraveling The Impact of Training Samples has been accepted by ICLR!

selected publications

  1. user-ideal-point-hack.png
    PAL: Pluralistic Alignment Framework for Learning from Heterogeneous Preferences
    Daiwei Chen, Yi Chen, Aniket Rege , and 1 more author
    2024
  2. learning-capacity.png
    Learning Capacity: A Measure of the Effective Dimensionality of a Model
    Daiwei Chen*, Weikai Chang*, and Pratik Chaudhari
    2023