NeurIPS 2022 Temporal Graph Learning Workshop
Important Information
NeurIPS virtual site (to follow livestream and recordings)
Workshop Date: Saturday, Dec. 3rd 2022; Room 399
Workshop Time Zone: New Orleans (GMT-5)
Workshop Date: Saturday, Dec. 3rd 2022; Room 399
Workshop Time Zone: New Orleans (GMT-5)
Contact Email: tglworkshop2022@gmail.com
Twitter: https://twitter.com/tgl_workshop
join us on slack
Theme
Graphs are prevalent in many diverse applications including Social networks, Natural Language Processing, Computer Vision, the World Wide Web, Political Networks, Computational finance, Recommender Systems and more. Graph machine learning algorithms have been successfully applied to various tasks, including node classification, link prediction and graph clustering. However, most methods assume that the underlying network is static thus limiting their applications to real-world networks which naturally evolve over time. On the one hand, temporal characteristics introduce substantial challenges compared to learning on static graphs. For example, in temporal graphs, the time dimension needs to be modelled jointly with graph features and structures. On the other hand, recent studies demonstrate that incorporating temporal information can improve the prediction power of graph learning methods thus creating new opportunities in applications such as recommendation system, event forecasting, fraud detection and more.
Investigation of temporal graphs provides the backbone of analysis of many different tasks including anomaly or fraud detection, disease modeling, recommendation systems, traffic forecasting, biology, social media, and many more. Hence, there has been a surge of interest in the development of temporal graph learning methods, from diverse domains spanning Machine Learning, Artificial Intelligence, Data Mining, Network Science, Public Health and more.
This workshop bridges the conversation among different areas such as temporal knowledge graph learning, graph anomaly detection, and graph representation learning. It aims to share understanding and techniques to facilitate the development of novel temporal graph learning methods. It also brings together researchers from both academia and industry and connects researchers from various fields aiming to span theories, methodologies, and applications.
Schedule
Keynote Talks
S. Mehran Kazemi: Learning and Reasoning with Temporal Knowledge Graphs
Google Research
Speaker Bio: Mehran Kazemi is a research scientist at Google Research, where he does fundamental and applied research on different areas of machine learning. His main areas of research include graph representation learning, time-series analysis, and natural language processing/understanding. Before joining Google Research, he was a research team lead at Borealis AI, a Royal Bank of Canada (RBC) institute for machine learning research. He received his Ph.D. and M.Sc. from the University of British Columbia working with prof. David Poole. Prior to that, he received his B.Sc. from Amirkabir University of Technology.
Bryan Hooi: Temporal Graph Learning: Some Challenges and Recent Directions
National University of Singapore
Speaker Bio: Bryan Hooi is an assistant professor in the School of Computing and the Institute of Data Science in National University of Singapore. He received his PhD degree in Machine Learning from Carnegie Mellon University, USA in 2019. His research interests include methods for learning from graphs and other complex or multimodal datasets, with the goal of developing efficient and practical approaches for applications ranging from fraud and scam detection, misinformation detection, to automatic monitoring of medical, traffic, and environmental sensor data.
Vikas Garg : Provably Powerful Temporal Graph Networks
Aalto University and YaiYai
Speaker Bio: Vikas Garg is an Assistant Professor of AI and Quantum Computing at Aalto University, and Cofounder and Chief Scientist at YaiYai. He received his PhD in Computer Science from MIT, and led research and engineering efforts during time with IBM Research (energy & wireless systems), Microsoft Research (edge computing), and Amazon A9 (e-commerce). His work has contributed to advancing multiple domains including drug discovery (the first graph-based deep learning model for protein design) and sustainable solutions (the first near-optimal method for integrating intermittent renewable energy into smart grids), and has been widely deployed in industry as well as public sectors. He has also served as a co-chair at the FCAI-IIT Conference on Deployable AI, Area Chair/Senior Program Committee member at premier AI/ML venues, BP Technologies Ventures Fellow at MIT, and researcher at MLPDS (a consortium of leading pharma companies).
Srijan Kumar: Temporal GNNs for Web Safety and Integrity
Georgia Institute of Technology
Speaker Bio: Srijan Kumar is an Assistant Professor at the College of Computing at Georgia Institute of Technology. He develops data science, machine learning, and AI solutions for the pressing challenges pertaining to the safety, integrity, and well-being of users, platforms, and communities in the cyber domain. He has pioneered the development of user models and network science tools to enhance the well-being and safety of users. His methods have been used in production at Flipkart (India’s largest e-commerce platform) and taught at graduate level courses worldwide. He has named to the Forbes 30 under 30 Class of 2022, named as a Kavli Fellow, and has received several awards including the Facebook Faculty Award, Adobe Faculty Award, ACM SIGKDD Doctoral Dissertation Award runner-up 2018, Larry S. Davis Doctoral Dissertation Award 2018, and 'best of' awards from WWW and ICDM. His research has been the subject of a documentary and covered in popular press, including CNN, The Wall Street Journal, Wired, and New York Magazine. He completed his postdoctoral training at Stanford University, received a Ph.D. in Computer Science from University of Maryland, College Park, and B.Tech. from Indian Institute of Technology, Kharagpur.
Pan Li: Representation learning for predicting temporal network evolution.
Purdue University
Speaker Bio: Pan Li has joined the Purdue CS department as an assistant professor since 2020 Fall. Before joining Purdue, Pan worked as a postdoc in the SNAP group at Stanford for one year, where he worked in the SNAP group led by Prof. Jure Leskovec. Before joining the SNAP group, Pan did his Ph.D. in Electrical and Computer Engineering at the University of Illinois Urbana - Champaign (2015 - 2019). His PhD advisor at UIUC was Prof. Olgica Milenkovic. At UIUC, he also worked with several wonderful collaborators including Prof. Niao He, Prof. Arya Mazumdar, Prof. Jiawei Han, Prof. David Gleich, etc. Before coming to UIUC, Pan Li received his M.S. degree in Electronic Engineering from Tsinghua University where his advisor was Prof. Xiqin Wang and he also worked with Prof. Huadong Meng and Prof. Yuan Shen. Before that, he got my B.S. degrees in both Physics and Electrical Engineering from Beijing Jiaotong University. He has spent two wonderful summers in Google and worked with several excellent researchers and engineers from Google AI, Google Cloud, and Google Ads. He is regularly invited as a reviewer for PNAS, PRMI, NeurIPS, ICML, ICLR, AISTATS, UAI, ISIT, AAAI, WWW, KDD, WSDM, SIGIR. He has received several awards such as JPMorgan Faculty Award and Ross-Lynn Faculty Award.
Panel
Aalto University and YaiYai
Purdue University
Georgia Institute of Technology
Imperial College London
Organizers
Farimah Poursafaei
McGill University/
Mila AI Institute
Accepted Papers
A Simple But Powerful Graph Encoder for Temporal Knowledge Graph Completion
Time-Evolving Conditional Character-centric Graphs for Movie Understanding
Long Hoang Dang, Thao Minh Le, Vuong Le, Tu Minh Phuong, Truyen Tran
Influencer Detection with Dynamic Graph Neural Networks
Correa Bahnsen, Bart Baesens, Monique Snoeck, Cristián Bravo
Learning Dynamic Graph Embeddings Using Random Walk With Temporal Backtracking
Chenghan Huang, Lili Wang, Xinyuan Cao, Weicheng Ma, Soroush Vosoughi
On the Evaluation of Methods for Temporal Knowledge Graph Forecasting
Julia Gastinger, Timo Sztyler, Anett Schuelke, Lokesh Sharma
Scalable Spatiotemporal Graph Neural Networks
Andrea Cini, Ivan Marisca, Filippo Maria Bianchi, Cesare Alippi
Victor Chomel, Nathanaël Cuvelle-Magar, Maziyar Panahi, David Chavalarias
Imperceptible Adversarial Attacks on Discrete-Time Dynamic Graph Models
Kartik Sharma, Rakshit Trivedi, Rohit Sridhar, Srijan Kumar
TTERGM: Social Theory-Driven Network Simulation
Yifan Huang, Clayton Thomas Barham, Eric Page, PK Douglas
Ali Behrouz, Margo Seltzer
Approximate Bayesian Computation for Panel Data with Signature Maximum Mean Discrepancies
Joel Dyer, John Fitzgerald, Bastian Rieck, Sebastian M Schmon
Program Committee
Aarash Feizi (McGill University/Mila AI Institute)
Anindya Mondal (University of Surrey)
Carl Yang (Emory University)
Chuxu Zhang (Brandeis University)
Dan Zhao (Yale University)
Derek Lim (MIT)
Dongkuan Xu (North Carolina State University)
Farimah Poursafaei (McGill University/Mila AI Institute)
Haoran Liu (Texas A&M)
Hejie Cui (Emory University)
Jacob Danovitch (McGill University/Mila AI Institute)
Jianan Zhao (U. de Montréal/Mila AI Institute)
Kartik Sharma (Georgia Institute of Technology)
Kellin Pelrine (McGill University/Mila AI Institute)
Kexin Huang (Stanford University)
Khaled mohammed Saifuddin (Georgia State University)
Ladislav Rampášek (U. de Montréal/Mila AI Institute)
Limei Wang (Texas A&M)
Lun Du (Microsoft Research Asia)
Haitao Mao (Michigan State University)
Mehwish Alam (FIZ-Karlsruhe)
Meng Qu (University of Montreal / Mila)
Mikhail Galkin (McGill University/Mila AI Institute)
Pratheeksha Nair (McGill University/Mila AI Institute)
Salvish Goomanee (College de France)
Shenyang Huang (McGill University/Mila AI Institute)
Vijay Prakash Dwivedi (Nanyang Technological University)
Yiwei Wang (National University of Singapore)
Yixin Liu (Monash University)
Yuanfu Lu (WeChat, Tencent)
Zepeng Zhang (ShanghaiTech University)
Call for Papers
Camera Ready Deadline: Nov. 23rd, 2022, Anywhere on Earth
Submission Deadline: Sep. 23rd, 2022, Anywhere on Earth (extended)
Accept/Reject Notification: Oct. 18th, 2022, Anywhere on Earth
We encourage researchers to submit their papers to this workshop on topics broadly related to the workshop themes above. We also welcome papers that present benchmark datasets, evaluation protocols, and challenges on relevant topics, including but not limited to:
Learning from temporal knowledge graphs, temporal hypergraphs and other types of evolving graphs
Data mining and visualization for temporal graphs
Benchmarking and evaluation of temporal graph learning methods
Novel temporal graph datasets
Challenges and solutions for large-scale training and deployment of temporal graph learning methods
Integration of temporal graphs with other fields such as computer vision, natural language processing, reinforcement learning, financial security, etc.
Question answering and common sense reasoning with evolving graph structure
Visual question answering with knowledge graphs
Theoretical analysis of temporal graphs, their properties and evolution
Temporal graph applications: anomaly detection, disease modeling, fraud detection, social media, traffic forecasting, recommendation systems, etc.
Examples of temporal graph representation learning across the natural and social sciences
As temporal graph learning is a rapidly evolving field, if you are unsure about the relevance of a topic, feel free to reach out to tglworkshop2022@gmail.com for clarification.
Authors can submit their papers through OpenReview. All submissions should use this template, and submissions should be in .pdf format.
The review process is double-blind, thus the papers should be anonymized appropriately. Previously published work is not accepted. This also includes papers accepted at the main NeurIPS conference this year.
Submissions should be no more than 8 pages (with unlimited pages for references and supplementary materials). We suggest authors only include minor details (e.g., hyperparameter settings) in the supplementary material. Shorter 4-page submissions presenting work in progress or discussing open problems and challenges in the domain of temporal graph learning are also welcome.
All accepted papers will be presented as posters during the workshop and the camera-ready version will be hosted on the workshop website (but not considered archival for resubmission purposes). In addition, four selected papers will be presented as spotlight talks during the workshop and one will receive the Best Paper Award. Authors of accepted papers will be asked to prepare a short video describing their work on SlidesLive which will also be hosted on the website.
Contact Email: tglworkshop2022@gmail.com
Twitter: https://twitter.com/tgl_workshop