Abstract:

Networks are a fundamental tool for understanding and modeling complex systems in physics, biology, neuro, and social sciences. Current network algorithms are almost exclusively focusing on first-order, or edge-based, structures in networks. However, what is missing from the picture are methods for analyzing higher-order organization of complex networks. We present a generalized framework for a network clustering and classification based on higher-order network connectivity patterns. This framework allows for identifying higher-order clusters in networks as well as for learning features of nodes in a network. Our framework scales to networks with billions of edges and provides mathematical guarantees on the optimality of obtained results. We apply our framework to networks from a variety of scientific domains with scales ranging from a few hundred to over one billion links.

Bio:

Jure Leskovec is an associate professor of Computer Science at Stanford University where he researches mining and modeling large social and information networks, their evolution, the diffusion of information, and its influence over these networks. In addition to his work at Stanford, Leskovec also serves as the chief scientist at Pinterest, focusing on machine learning problems for the popular social media platform. 

Computation over massive data is at the heart of Leskovec’s research and has applications in computer science, social sciences, economics, marketing, and health care. This research has won several awards, including a Lagrange Prize, a Microsoft Research Faculty Fellowship, the Alfred P. Sloan Fellowship, and numerous best paper awards. 

Leskovec received his bachelor’s degree in computer science from the University of Ljubljana, Slovenia, and his Ph.D. in machine learning from Carnegie Mellon University. He conducted his postdoctoral training at Cornell University.