Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs (2017-04-10T00:00:00.000000Z)

TL;DR

This work generalizes the convolution operator from regular grids to arbitrary graphs while avoiding the spectral domain, which allows us to handle graphs of varying size and connectivity.

Abstract

A number of problems can be formulated as prediction on graph-structured data. In this work, we generalize the convolution operator from regular grids to arbitrary graphs while avoiding the spectral domain, which allows us to handle graphs of varying size and connectivity. To move beyond a simple diffusion, filter weights are conditioned on the specific edge labels in the neighborhood of a vertex. Together with the proper choice of graph coarsening, we explore constructing deep neural networks for graph classification. In particular, we demonstrate the generality of our formulation in point cloud classification, where we set the new state of the art, and on a graph classification dataset, where we outperform other deep learning approaches.

Authors

N. Komodakis

8 papers

M. Simonovsky

1 papers

TL;DR

Abstract

Authors

References45 items

Point cloud labeling using 3D Convolutional Neural Network

Graph Based Convolutional Neural Network

Semi-Supervised Classification with Graph Convolutional Networks

Layer Normalization

Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering

Dynamic Filter Networks

Learning Convolutional Neural Networks for Graphs

Orientation-boosted Voxel Nets for 3D Object Recognition

Volumetric and Multi-view CNNs for Object Classification on 3D Data

Discriminative Embeddings of Latent Variable Models for Structured Data

Deep Residual Learning for Image Recognition

All you need is a good init

Gated Graph Sequence Neural Networks

Diffusion-Convolutional Neural Networks

Convolutional Networks on Graphs for Learning Molecular Fingerprints

VoxNet: A 3D Convolutional Neural Network for real-time object recognition

Deep Graph Kernels

Multi-view Convolutional Neural Networks for 3D Shape Recognition

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Geodesic Convolutional Neural Networks on Riemannian Manifolds

Fractional Max-Pooling

Convolutional neural networks at constrained time cost

GSPBOX: A toolbox for signal processing on graphs

3D ShapeNets for 2.5D Object Recognition and Next-Best-View Prediction

Performance of global descriptors for velodyne-based urban object recognition

Exact solutions to the nonlinear dynamics of learning in deep linear neural networks

Spectral Networks and Locally Connected Networks on Graphs

A Multiscale Pyramid Transform for Graph Signals

Advanced Coarsening Schemes for Graph Partitioning

3D is here: Point Cloud Library (PCL)

Kron Reduction of Graphs With Applications to Electrical Networks

Weisfeiler-Lehman Graph Kernels

Graph sparsification by effective resistances

Comparison of descriptor spaces for chemical compound retrieval and classification

Shortest-path kernels on graphs

Distinguishing enzyme structures from non-enzymes without alignments.

Structure-activity relationship of mutagenic aromatic and heteroaromatic nitro compounds. Correlation with molecular orbital energies and hydrophobicity.

Deep Learning

I and J

The Graph Neural Network Model

GradientBased Learning Applied to Document Recognition

Gradient-based learning applied to document recognition

Un-supervised feature learning for classiﬁcation of outdoor 3d scans

(88.33%) to the same reason

Graph Sparsiﬁcation by Effective Resistances

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names