Title: Graph Representation Learning and Applications to Drug Discovery
Jian Tang (HEC Montreal)
October 20th, 12-1pmÂ
Zoom Link:Â https/mcgill.zoom.us/j/91589192037
Abstract: Graphs, a general type of data structures for capturing interconnected objects, are ubiquitous in a variety of disciplines and domains. In this talk, I will introduce our work on graph representation learning and applications to drug discovery. In the first part, I will introduce our work on graph embeddings including learning node representations (LINE, WWW’15), extremely low-dimensional node representation learning for graph and high-dimensional data visualization (LargeVis, WWW’16), knowledge graph embedding (RotatE, ICLR’19), and a general and high-performance graph embedding system (GraphVite, WWW’19, ). In the second part, I will introduce our recent work on graph representation learning for drug discovery including an unsupervised and semi-supervised approach for learning graph representations for molecule properties prediction and a new generative model for molecular graph generation and optimization.
