Normalizing flows provide a general mechanism for defining expressive probability
distributions, only requiring the specification of a (usually simple) base distribution and a …

The conjoining of dynamical systems and deep learning has become a topic of great
interest. In particular, neural differential equations (NDEs) demonstrate that neural networks …

Contemporary neural networks still fall short of human-level generalization, which extends
far beyond our direct experiences. In this paper, we argue that the underlying cause for this …

We introduce a new family of deep neural network models. Instead of specifying a discrete
sequence of hidden layers, we parameterize the derivative of the hidden state using a …

This paper builds on the connection between graph neural networks and traditional
dynamical systems. We propose continuous graph neural networks (CGNN), which …

Graph neural networks (GNN) are powerful tools in many web research problems. However,
existing GNNs are not fully suitable for many real-world web applications. For example, over …

Graph convolutional networks (GCNs) are discriminative models that directly model the
class posterior $ p (y|\mathbf {x}) $ for semi-supervised classification of graph data. While …

We exploit minimally faithful inversion of graphical model structures to specify sparse
continuous normalizing flows (CNFs) for amortized inference. We find that the sparsity of this …

Recently, many works have been proposed for traffic forecasting to improve people's daily
lives. Although these works have achieved good predictive performance, they have three …

Graph Convolution Networks (GCNs) are widely considered state-of-the-art for collaborative
filtering. Although several GCN-based methods have been proposed and achieved state-of …