Zhitao Ying, Ph.D., Stanford University; B.S., Duke University. Joined Yale Faculty July 2022.
My research spans 3 broad areas: deep learning for graphs, geometric representation learning, and real-world applications with relational reasoning and modeling. In the past, I created many widely used GNN algorithms such as GraphSAGE, PinSAGE and GNNExplainer.
In addition, I have worked on a variety of applications of graph learning in physical simulations, social networks, knowledge graphs and biology. I developed the first billion-scale graph embedding services at Pinterest, and the graph-based anomaly detection algorithm at Amazon.
Prospective students:
I’m looking for Ph.D. students who are passionate about pushing the frontiers of GNNs and geometric deep learning, through novel modeling of real-world problems and data, theoretical study of GNN and manifolds, new capabilities for graphs (e.g. autoML, explainability and fairness). Students interested in intersections of graph learning and applications in social sciences, natural sciences, medicine and industrial AI platforms are also welcome to reach out.
See Google Scholar profile for the entire list of publications.
- Modeling Heterogeneous Hierarchies with Relation-specific Hyperbolic Cones, NeurIPS 2021
- Neural Distance Embeddings for Biological Sequences, NeurIPS 2021
- Bipartite Dynamic Representations for Abuse Detection, KDD 2021
- Identity-aware Graph Neural Networks, AAAI 2021
- (Ph.D. Thesis) Towards Expressive and Scalable Deep Representation Learning for Graphs, 2021
- Multi-hop Attention Graph Neural Network, IJCAI 2021
- Learning to Simulate Complex Physics with Graph Networks, ICML 2020
- Redundancy-free Computation for Graph Neural Networks, KDD 2020
- Design space for graph neural networks, NeurIPS 2020
- Neural Execution of Graph Algorithms, ICLR 2020
- Gnnexplainer: Generating Explanations for Graph Neural Networks, NeurIPS 2019
- Hyperbolic Graph Convolutional Neural Networks, NeurIPS 2019
- Position-aware Graph Neural Networks, ICML 2019
- Graph Convolutional Neural Networks for Web-scale Recommender Systems, KDD 2018
- Hierarchical Graph Representation Learning with Differentiable Pooling, NeurIPS 2018
- Graph Convolutional Policy Network for Goal-directed Molecular Graph Generation, NeurIPS 2018
- Graphrnn: Generating Realistic Graphs with Deep Auto-regressive Models, ICML 2018
- Inductive Representation Learning on Large Graphs, NeurIPS 2017
- Representation learning on graphs: Methods and applications, IEEE Data Engineering Bulletin 2017
-
Baidu Scholarship 2019
-
AC for LoG 2022 Conference
-
Co-Lead of PyTorch Geometric Library for GNNs
-
Lead organizer of SimDL Workshop at ICLR 2021
-
Lead organizer of Stanford Graph Learning Workshop 2021
-
Co-organizer of GRL+ Workshop at ICML 2020