Weilin Cong (丛炜霖)

alt text 

PhD, The Pennsylvania State University
E-mail: congweilin95@gmail.com / weilin@psu.edu

About me

I am a 4th year Ph.D. candidate (since Fall 2019) of Computer Science and Engineering at The Pennsylvania State University, under the supervision of Prof. Mehrdad Mahdavi. I received my B.S. degree from Beijing institute of Technology.

Besides, my last name Cong (simplified 丛 / traditional 叢) is pronounced as ts-oh-ng in Pinyin 🧐

Research Interests

My research focuses on both the fundamental problems in graph representation learning (including optimization, generalization, and expressive power) and model architecture design:
  • Optimization: scaling graph neural network training by sampling, distributed graph neural network training, and efficient graph representation unlearning for privacy;

  • Generalization and expressive power: understanding why GNN suffer from performance degradation when the depth go deeper;

  • Model architecture design: design a Graph Transformer network that can efficiently solve the dynamic graph reasoning problem.

Please refer to the publications and most recent updates for more details.

Most recent update (06/11/2022)

Our most recent work on graph representation unlearning is available online. Our goal is to efficiently remove the effect of a set of nodes on the pre-trained GNN model. [Slides] [Video]
  • In GraphEditor, we first formulate GNNs as an alternative problem with closed-form solution, then we can edit the weight parameters based on the change of graph structure.

  • right
  • In Projector, based on our observation that the linear GNN's weight parameters (trained with logistic regression) are in the span of all node features, we propose to unlearn by projecting the original weight paramters into a subspace that is irrelevant to the deleted node features.

  • right

Working Experience

  • Meta AI Research, Menlo Park, CA

    Research intern, May.2022 - Aug.2022

    Supervisor: Si Zhang

    Topic: Dynamic graph reasoning using Transformers

  • Meta AI Research, Remote, US

    Research intern, Jun.2021 - Aug.2021

    Supervisor: Yanhong Wu

    Topic: Dynamic graph reasoning using Transformers