drawing drawing


Machine Learning on graphs, and more generally non-euclidean structured data, has recently emerged as a primary citizen of the machine learning world, with many applications in community detection, recommender systems, natural language processing, molecule classification, protein interface prediction, quantum chemistry, epidemiology, combinatorial optimization… and so on. In particular, deep models such as Graph Neural Networks have become very popular tools, with their successes, limitations, and a plethora of open questions.

This day aims to bring together researchers and students to discuss recent advances in Graph Machine Learning, both theoretical and practical. Topics include, but are not limited to:

  • Graph Neural Networks, graph kernels
  • Statistics on graphs, random graphs
  • Graph Signal Processing

Invited speakers

Call for participation

The day will include short (30-min) presentations and potentially a poster session. Interested participants should send the organizers a 1-page pdf abstract that includes the authors’ names and affiliations before January 21st. Students are particularly encouraged to participate.