Research

My research focuses on bridging the gap between purely academic research and real-world applications of Graph Neural Networks. In the real-world, graphs are often extremely large, dynamic, directed, and have nodes with partially missing features. This raises the following research questions:

  • Scalability: How can we scale GNNs to handle billion-node graphs and beyond?
  • Dynamic Graphs: How can we learn from graphs that change over time, such as social or transaction networks?
  • Directionality: How can we learn from graphs that have directed edges, such as transportation or citation networks?
  • Missing Node Features: How can we apply GNNs to graphs where we only observe a subset of features for each node?

Through my research, I aim to develop solutions to these challenges and enable GNNs to be used more widely and effectively in real-world applications.

Publications

Below is a list of my publications in reversed chronological order.


2023

  1. Temporal Graph Benchmark for Machine Learning on Temporal Graphs
    Shenyang Huang, Farimah Poursafaei, Jacob Danovitch, Matthias Fey, Weihua Hu,  Emanuele Rossi, Jure Leskovec, Michael M. Bronstein, Guillaume Rabusseau, and Reihaneh Rabbany
    Advances in Neural Information Processing Systems 2023
    arXiv Code
  2. Edge Directionality Improves Learning on Heterophilic Graphs
    Emanuele Rossi, Bertrand Charpentier, Francesco Di Giovanni, Fabrizio Frasca, Stephan Günnemann, and Michael M. Bronstein
    arXiv 2023
    arXiv Blog Code

2022

  1. Graph Neural Networks for Link Prediction with Subgraph Sketching
    Ben Chamberlain, Sergey Shirobokov,  Emanuele Rossi, Fabrizio Frasca, Thomas Markovich, Nils Hammerla, Michael M. Bronstein, and Max Hansmire
    International Conference on Learning Representations (ICLR) 2022
    arXiv Code
  2. Provably Efficient Causal Model-Based Reinforcement Learning for Systematic Generalization
    Mirco Mutti, Riccardo De Santi,  Emanuele Rossi, Juan Felipe Calderon, Michael M. Bronstein, and Marcello Restelli
    Proceedings of the AAAI Conference on Artificial Intelligence 2022
    arXiv
  3. Learning to Infer Structures of Network Games
    Emanuele Rossi, Federico Monti, Yan Leng, Michael M. Bronstein, and Xiaowen Dong
    Proceedings of the 39th International Conference on Machine Learning, ICML 2022
    arXiv Blog
  4. On the Unreasonable Effectiveness of Feature propagation in Learning on Graphs with Missing Node Features
    Emanuele Rossi, Henry Kenlay, Maria I. Gorinova, Ben Chamberlain, Xiaowen Dong, and Michael M. Bronstein
    Learning on Graphs Conference 2022
    arXiv Blog Code Slides

2021

  1. GRAND: Graph Neural Diffusion
    Ben Chamberlain, James Rowbottom, Maria I. Gorinova, Stefan D. Webb,  Emanuele Rossi, and Michael M. Bronstein
    Proceedings of the 38th International Conference on Machine Learning, ICML 2021
    arXiv Blog Code

2020

  1. Tuning Word2vec for Large Scale Recommendation Systems
    Ben Chamberlain,  Emanuele Rossi, Dan Shiebler, Suvash Sedhain, and Michael Bronstein
    RecSys - 14th ACM Conference on Recommender Systems 2020
    arXiv
  2. Temporal Graph Networks for Deep Learning on Dynamic Graphs
    Emanuele Rossi, Ben Chamberlain, Fabrizio Frasca, Davide Eynard, Federico Monti, and Michael Bronstein
    ICML Workshop on Graph Representation Learning 2020
    arXiv Blog Code Slides
  3. SIGN: Scalable Inception Graph Neural Networks
    Emanuele Rossi, Fabrizio Frasca, Ben Chamberlain, Davide Eynard, Michael Bronstein, and Federico Monti
    ICML Workshop on Graph Representation Learning 2020
    arXiv Blog Code

2019

  1. ncRNA Classification with Graph Convolutional Networks
    Emanuele Rossi, Federico Monti, Michael Bronstein, and Pietro Liò
    KDD Workshop on Deep Learning on Graphs 2019
    arXiv Code