Abstract We propose a framework for the definition of neural models for graphs that do not rely on backpropagation for training, thus making learning more biologically plausible and amenable to parallel implementation. Our proposed framework is inspired by Gated Linear Networks and allows the adoption of multiple graph convolutions. Specifically, each neuron is defined as a set of graph convolution filters (weight vectors) and a gating mechanism that, given a node and its topological context, generates the weight vector to use for processing the node’s attributes. Two different graph processing schemes are studied, i.e., a message-passing aggregation scheme where the gating mechanism is embedded directly into the graph convolution, and a multi-resolution one where neighboring nodes at different topological distances are jointly processed by a single graph convolution layer. We also compare the effectiveness of different alternatives for defining the context function of a node, i.e., based on hyperplanes or on prototypes, and using a soft or hard-gating mechanism. We propose a unified theoretical framework allowing us to theoretically characterize the proposed models’ expressiveness. We experimentally evaluate our backpropagation-free graph convolutional neural models on commonly adopted node classification datasets and show competitive performances compared to the backpropagation-based counterparts.

Many neural networks for graphs are based on the graph convolution (GC) operator, proposed more than a decade ago. Since then, many alternative definitions have been proposed, which tend to add complexity (and nonlinearity) to the model. Recently, however, a simplified GC operator, dubbed simple graph convolution (SGC), which aims to remove nonlinearities was proposed. Motivated by the good results reached by this simpler model, in this article we propose, analyze, and compare simple graph convolution operators of increasing complexity that rely on linear transformations or controlled nonlinearities, and that can be implemented in single-layer graph convolutional networks (GCNs). Their computational expressiveness is characterized as well. We show that the predictive performance of the proposed GC operators is competitive with the ones of other widely adopted models on the considered node classification benchmark datasets.

We propose a class of neural models for graphs that do not rely on backpropagation for training, thus making learning more biologically plausible and amenable to parallel implementation in hardware. The base component of our architecture is a generalization of Gated Linear Networks which allows the adoption of multiple graph convolutions. Specifically, each neuron is defined as a set of graph convolution filters (weight vectors) and a gating mechanism that, given a node and its topological context, selects the weight vector to use for processing the node’s attributes. Two different graph processing schemes are studied, i.e., a message-passing aggregation scheme where the gating mechanism is embedded directly into the graph convolution, and a multi-resolution one where neighbouring nodes at different topological distances are jointly processed by a single graph convolution layer. We also compare the effectiveness of different alternatives for defining the context function of a node, i.e., based on hyper-planes or on prototypes. A theoretical result on the expressiveness of the proposed models is also reported. We experimented our backpropagation-free graph convolutional neural architectures on commonly adopted node classification datasets, and show competitive performances compared to the backpropagation-based counterparts.

We present a flexible framework for uncertainty principles in spectral graph theory. In this framework, general filter functions modeling the spatial and spectral localization of a graph signal can be incorporated. It merges several existing uncertainty relations on graphs, among others the Landau-Pollak principle describing the joint admissibility region of two projection operators, and uncertainty relations based on spectral and spatial spreads. Using theoretical and computational aspects of the numerical range of matrices, we are able to characterize and illustrate the shapes of the uncertainty curves and to study the space-frequency localization of signals inside the admissibility regions.

Magnetic Particle Imaging (MPI) is an emerging technology in the field of (pre)clinical imaging. The acquisition of a particle signal is realized along specific sampling trajectories covering a defined field of view (FOV). In a system matrix (SM) based reconstruction procedure, the commonly used acquisition path in MPI is a Lissajous trajectory. Such a trajectory features an inhomogeneous coverage of the FOV, i.e. the center region is sampled less dense than the regions towards the edges of the FOV. Conventionally, the respective SM acquisition and the subsequent reconstruction do not reflect this inhomogeneous coverage. Instead, they are performed on an equispaced grid. The objective of this work is to introduce a sampling grid that inherently features the aforementioned inhomogeneity by using node points of Lissajous trajectories. Paired with a tailored polynomial interpolation of the reconstructed MPI signal, the entire image can be recovered. It is the first time that such a trajectory related non-equispaced grid is used for image reconstruction on simulated and measured MPI data and it is shown that the number of sampling positions can be reduced, while the spatial resolution remains constant.