End-to-end learning on graphs with graph convolutional networks

Neural networks on graphs have gained renewed interest in the machine learning community. Recent results have shown that end-to-end trainable neural network models that operate directly on graphs can challenge well-established classical approaches, such as kernel-based methods or methods that rely on graph embeddings (e.g. DeepWalk). In this talk, I will motivate such an approach from an analogy to traditional convolutional neural networks and introduce our recent variant of graph convolutional networks (GCNs) that achieves promising results on a number of semi-supervised node classification tasks. I will further introduce two extensions to this basic framework, namely: graph auto-encoders and relational GCNs. While graph auto-encoders provide a novel way of approaching problems like link prediction and clustering, relational GCNs allow for efficient modeling of directed, relational graphs, such as knowledge bases (e.g. Freebase).

Short bio

Thomas Kipf is a second-year PhD student at the University of Amsterdam, advised by Prof. Max Welling. His research focuses on large-scale inference for structured data, including topics such as semi-supervised learning, reasoning, and multi-agent reinforcement learning. During his earlier studies in Physics, he has had exposure to a number of fields, and—after a short interlude in Neuroscience-related research at the Max Planck Institute for Brain Research—eventually developed a deep interest in machine learning and AI.