Bio

I am a final year Ph.D. student at Mila and McGill University , supervised by Professor Reihaneh Rabbany and Professor Guillaume Rabusseau . Previously I obtained an Honours in Computer Science from McGill University in 2019. I have a broad interest in temporal graph neural networks, graph transformers, graph neural networks and spectral methods. My research focuses on machine learning models for complex and evolving networks in the real world, referred to as Temporal Graph Learning (TGL). I also actively engage in building the TGL community by organizing the TGL reading group and two editions of the TGL workshop @ NeurIPS 2022 / 2023. Through my research, I aim to answer the following questions:

• How to design more scalable, expressive and powerful temporal graph methods?
• How to properly evaluate temporal graph methods based on real world considerations?
• How to deploy TGL methods for applications such as disease modeling, anomaly detection and forecasting?

News!

[2024/11] Excited to share that our work, UTG: Towards a Unified View of Snapshot and Event Based Models for Temporal Graphs has been accepted at LOG 2024 Conference. See you at LOG conference. Thanks to my amazing co-authors: Farimah Poursafaei, Reihaneh Rabbany, Guillaume Rabusseau and Emanuele Rossi.

[2024/10] Excited to share that our work TGB 2.0: A Benchmark for Learning on Temporal Knowledge Graphs and Heterogeneous Graphs has been accepted at NeurIPS 2024 Datasets and Benchmark Track. Thanks to my amazing co-authors and see you at NeurIPS 2024.

[2024/10] Excited to share that our preprint Temporal Graph Rewiring with Expander Graphs has achieved state-of-the-art results on four TGB datasets, demonstrating the potential of graph rewiring on temporal graphs. Thanks to my co-authors Katarina Petrović, Farimah Poursafaei and Petar Veličković.

[2024/06] Excited to annouce our paper MiniMol: A Parameter-Efficient Foundation Model for Molecular Learning is selected as a spotlight poster at ICML 2024 Workshop on Efficient and Accessible Foundation Models for Biological Discovery. Thanks to my amazing collaborators at GraphCore, RWTH Aachen University, New Jersey Institute of Technology and Mila.

[2024/05] Excited to announce our paper Static graph approximations of dynamic contact networks for epidemic forecasting is accepted at Scientific Reports. Thanks to my amazing co-authors Razieh Shirzadkhani, Abby Leung and Reihaneh Rabbany.

[2024/01] Excited to annouce two papers accepted at ICLR 2024, Graphpulse: Topological representations for temporal graph property prediction and Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task Datasets. Thanks to all my amazing collaborators!

Publications

2024

Gastinger, J.* Huang, S.*, Galkin, M., Loghmani, E., Parviz, A., Poursafaei, F., Danovitch, J., Rossi, E., Koutis, I., Stuckenschmidt, H., Rabbany, R., and Rabusseau, G. TGB 2.0: A Benchmark for Learning on Temporal Knowledge Graphs and Heterogeneous Graphs (NeurIPS 2024 Datasets and Benchmark Track)

Huang, S.*, Poursafaei, F.*, Rabbany, R., Rabusseau, G. and Rossi, E. UTG: Towards a Unified View of Snapshot and Event Based Models for Temporal Graphs (LOG 2024)

Shirzadkhani, R.*, Ngo, T. G. B.*, Shamsi, K.* Huang, S., Poursafaei, F., Azad, P., Rabbany, R., Coskunuzer, B., Rabusseau, G., and Akcora, G. C. Towards Neural Scaling Laws for Foundation Models on Temporal Graphs (preprint)

Petrović., K., Huang, S., Poursafaei, F., and Veličković, P. Temporal Graph Rewiring with Expander Graphs (ICML 2024 Workshop on Geometry-grounded Representation Learning and Generative Modeling)

Kläser, K., Banaszewski, B., Maddrell-Mander, S., McLean, C., Müller, L., Parviz, A., Huang, S. and Fitzgibbon, A. MiniMol: A Parameter-Efficient Foundation Model for Molecular Learning (Spotlight Poster at ICML 2024 Workshop on Efficient and Accessible Foundation Models for Biological Discovery)

Shirzadkhani, R., Huang, S., Leung, A. and Rabbany, R. Static graph approximations of dynamic contact networks for epidemic forecasting (Scientific Reports)

Shamsi, K., Poursafaei, F., Huang, S., Ngo, B., Coskunuzer, B. and Akcora, C. Graphpulse: Topological representations for temporal graph property prediction (ICLR 2024)

Beaini, D., Huang, S., Cunha, J.A., Li, Z., Moisescu-Pareja, G., Dymov, O., Maddrell-Mander, S., McLean, C., Wenkel, F., Müller, L., Mohamud, J., Parviz, A., Craig, M., Koziarski, M., Lu, J., Zhu, Z., Gabellini, C., Klaser, K., Dean, J., Wognum, C., Sypetkowski, M., Rabusseau, G., Rabbany, R., Tang, J., Morris, C., Koutis, I., Ravanelli, M., Wolf, G., Tossou, P., Mary, H., Bois, T., Fitzgibbon, A., Banaszewski, B., Martin, C., and Masters, D. Towards Foundational Models For Molecular Learning on Large-scale Multi-task Datasets (ICLR 2024)

Shirzadkhani, R., Huang, S.,, Kooshafar, E., Rabbany, R., and Poursafaei, F. Temporal Graph Analysis with TGX (WSDM 2024 Demo Track)

2023

Huang, S.*,, Poursafaei, F.*, Danovitch, J., Fey, M., Hu, W., Rossi, E., Leskovec, J., Bronstein, M., Rabusseau G. and Rabbany R. Temporal Graph Benchmark for Machine Learning on Temporal Graphs (NeurIPS 2023 Datasets and Benchmarks Track)

Huang, S.,, Danovitch, J., Rabusseau G., Rabbany R. Fast and Attributed Change Detection on Dynamic Graphs with Density of States (PAKDD 2023)

Huang, S., Coulombe, S., Hitti, Y., Rabbany, R., Rabusseau, G. Laplacian Change Point Detection for Single and Multi-view Dynamic Graphs (ACM Transactions on Knowledge Discovery from Data, TKDD)

Masters, D., Dean, J. Klaser, K., Li, Z., Mander, S., Sanders, A., Helal, H., Beker, D., Fitzgibbon, A., Huang, S., Rampášek, L., Beaini, D. GPS++: Reviving the Art of Message Passing for Molecular Property Prediction (TMLR)

Pupneja, Y., Zou, J., Lévy, S. Huang, S., Understanding Opinions Towards Climate Change on Social Media (NeurIPS 2023 Workshop: Tackling Climate Change with Machine Learning)

Jiang, L., Zhang, C., Poursafaei, F., Huang, S., Towards Temporal Edge Regression: A Case Study on Agriculture Trade Between Nations (preprint)

2022

Poursafaei, F.*, Huang, S.*, Pelrine, K., Rabbany, R. Towards Better Evaluation for Dynamic Link Prediction (NeurIPS 2022 Datasets and Benchmarks Track)

2021

Huang, S., Wang, K., Rabusseau, G., & Makhzani, A. Few Shot Image Generation via Implicit Autoencoding of Support Sets 5th Workshop on Meta-Learning at NeurIPS 2021

Huang, S., Rabusseau, G. & Rabbany, R. Scalable Change Point Detection for Dynamic Graphs 6th Outlier Detection and Description Workshop at KDD 2021

Huang, S., François-Lavet, V., & Rabusseau, G. Understanding Capacity Saturation in Incremental Learning. Canadian Conference on Artificial Intelligence 2021

Ding, X., Huang, S., Leung, A., Rabbany, R. Incorporating dynamic flight network in SEIR to model mobility between populations. Applied Network Science, Special issue on Epidemics Dynamics & Control on Networks

2020

Huang, S., Hitti, Y., Rabusseau, G. & Rabbany, R. Laplacian Change Point Detection for Dynamic Graphs. (KDD 2020)

Leung, A., Ding, X., Huang, S., Rabbany, R. Contact Graph Epidemic Modelling of COVID-19 for Transmission and Intervention Strategies.

Alletto, S., Huang, S., François-Lavet, V., Nakata, Y., & Rabusseau, G. RandomNet: Towards Fully Automatic Neural Architecture Design for Multimodal Learning. AAAI 2020 Meta-Eval Workshop

2019

Huang, S., François-Lavet, V., & Rabusseau, G. Neural Architecture Search for Class-incremental Learning
(previous version of "Understanding Capacity Saturation in Incremental Learning")

2018

Huang, S., François-Lavet, V., Rabusseau, G. & Pineau, J. Exploring Continual Learning Using Incremental Architecture Search NeuIPS Continual Learning Workshop 2018.

Blogs

Teaching

Winter 2024, Mentor Geometric Deep Learning L65, University of Cambridge

Winter 2023, Mentor, Representation Learning on Graphs and Networks L45, University of Cambridge

Fall 2022, Guest Lecturer, Anomaly Detection for Dynamic Graphs (updated slides), COMP 599 Network Science, McGill University

Fall 2021, Guest Lecturer, Anomaly Detection for Dynamic Graphs, COMP 599 Network Science, McGill University

Fall 2021, TA, COMP 599 Network Science, McGill University

Winter 2020, TA, COMP 250, Introduction to Computer Science, McGill University

Fall 2019, TA, COMP 202, Introduction to Programming, McGill University

Services

Organizer of the weekly Temporal Graph Reading Group

Organizer of the Temporal Graph Learning Community Slack, see here to join

Organization Chair for Temporal Graph Learning Workshop@NeurIPS 2023

Organization Chair for Temporal Graph Learning Workshop@NeurIPS 2022

ICLR 2025 Reviewer

NeurIPS 2024 Reviewer

NeurIPS 2024 Datasets and Benchmarks Track Reviewer

NeurIPS 2023 Datasets and Benchmarks Track Reviewer

NeurIPS 2022 Datasets and Benchmarks Track Reviewer

Scientific Reports Reviewer 2024, 2 papers

Reviewer for Transactions on Machine Learning Research (TMLR) journal 2023

KDD 2021 External Reviewer

IEEE Transactions on Neural Networks and Learning Systems Reviewer 2021

ECML PKDD 2020 Program Committee Member

Awards and Scholarships

NSERC Postgraduate Scholarships-Doctoral (PGS D) Award, 2022-2025

Fonds de recherche du Québec – Nature et Technologies (FRQNT) Doctoral Award, 2022-2026

Scientist in Residence Program, Valence Labs and Mila, 2023

McGill Graduate Research Enhancement and Travel Awards (GREAT awards), 2023

Mitacs Accelerate Award, 2022

NSERC Undergraduate Student Research Awards, 2018

McGill Undergraduate Computer Science Research Award, 2nd Place Winner, 2018

McGill Physics Hackathon, 2nd Place Winner, 2017

NSERC Undergraduate Student Research Awards, 2016

Mentorship

2024 April-May, Research Intern, Julia Gastinger, TGB 2.0

2024, Research Assistant, Razieh Shirzadkhani

2024, Project Mentor, Qianyi Liu and Zak Buzzard, Better time encoding models in temporal GNNs.

2024, Project Mentor, Dobromir Marinov and Lauren R. Wilkes, Adversarial Attacks on continuous-time financial networks

2024, Project Mentor, Kiril Bikov and Riccardo Conci, Looking into the past: exploring time dependencies beyond exponential decay

2023-2024, Research Intern, Shahrad Mohammadzadeh, better temporal graph learning methods.

2022-2023, Research Assistant, Razieh Shirzadkhani, Disease Modeling with Dynamic Graphs & TGX package

2023, Project Mentor, Lekang Jiang and Caiqi Zhang, Towards Temporal Edge Regression: A Case Study on Agriculture Trade Between Nations

2023, Project Mentor, Xiangjian Jiang and Yanyi Pu, Exploring Time Granularity on Temporal Graphs for Dynamic Link Prediction in Real-world Networks

2022, Research Assistant, Abu bakar Daud, pypi package for Towards better dynamic link prediction