The awarded paper, “No PANE, No Gain: Scaling Attributed Network Embedding in a Single Server”, was co-authored with his former PhD student Dr Yang Renchi (Class of 2021), and collaborators at the National University of Singapore, Hamida Bin Khalifa University, and the Hong Kong Polytechnique University.
The work was previously awarded "Best Research Paper Award" in VLDB 2021. It addresses the lack of scalability of existing attributed network embedding (ANE) techniques by proposing PANE, an effective and scalable approach for massive graphs in a single server that achieves state-of-the-art result quality on multiple benchmark datasets for two common prediction tasks: link prediction and node classification. Under the hood, PANE takes inspiration from well-established data management techniques to scale up ANE in a single server. Specifically, it exploits a carefully formulated problem based on a novel random walk model, a highly efficient solver, and non-trivial parallelization by utilizing modern multi-core CPUs. Extensive experiments demonstrate that PANE consistently outperforms all existing methods in terms of result quality, while being orders of magnitude faster.
The potential outgrowths of the work are broader. It is well-known that ML/DL techniques are highly energy-intensive with potentially large environmental impacts. The proposed solution in the paper paves the way towards sustainable graph analytics. For example, on the largest Microsoft Academic Graph, PANE is the only viable solution that computes the embeddings on a single server within 12 hours instead of demanding an energy-hungry multi-GPU cluster that may run for days to address the problem.
The SIGMOD Research Highlight Award is a highly selective and prestigious award that aims to showcase a set of research projects that exemplify core database research. These projects address an important problem, represent a definitive milestone in solving
the problem, and have the potential of significant impact. The initiative of the SIGMOD Research Highlights also aims to make the selected works widely known in the database community, to its industry partners as well as the broader ACM community.
The ACM Special Interest Group on Management of Data (ACM SIGMOD) is concerned with the principles, techniques and applications of database management systems and data management technology. Its members include software developers, academic and industrial researchers, practitioners, users, and students. SIGMOD sponsors the annual SIGMOD/PODS conference, which is one of the most important and selective in the field.