Yujun's research lies at the intersection of machine learning (ML) and network science. Her work provides theoretical understanding and empirical practices to ML models, with application to complex real-world networks. She is particularly interested in (1) fundamental principles in designing more expressive and generalizable graph-based ML models and (2) useful practices of applying graph-based models to various domains, such as neuroscience, and program understanding.
- BS Southeast University, Nanjing China
- MS University of Michigan-Ann Arbor
- PhD University of Michigan-Ann Arbor
Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks. Yujun Yan, Milad Hashemi, Kevin Swersky, Yaoqing Yang, Danai Koutra. The IEEE International Conference on Data Mining (ICDM) 2022.
Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs. Jiong Zhu, Yujun Yan, Lingxiao Zhao, Mark Heimann, Leman Akoglu, Danai Koutra. Conference on Neural Information Processing Systems (NeurIPS) 2020.
Neural Execution Engines: Learning to Execute Subroutines. Yujun Yan, Kevin Swersky, Danai Koutra, Parthasarathy Ranganathan, Milad Hashemi. Conference on Neural Information Processing Systems (NeurIPS) 2020.