Building Robust Graph Embeddings for Massive Real World Graphs


Graph based neural networks have garnered a lot of attention over the past few years especially in Search and Recommendation technology. Large global web companies like Amazon, Facebook, LinkedIn and Google use graph based models both in production and offline to develop robust representations of their existing knowledge graph systems . However, large scale graph modeling brings in a host of new challenges both on the machine learning architecture front and the distributed computing front. Large real world graphs are noisy, have a power-law distribution, contain high-behavior hub nodes which cause load imbalance during training/inference on the one hand and non-behavioural nodes with cold-start issues on the other. In this talk, I’ll go over the approaches to overcome challenges in large scale GNN modeling and present methods to scale and productionize GNN based models. I’ll briefly go over our recent publications in WWW and ICLR which provide research directions in this area.


Bio Coming Soon!

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google