Graph Node-Feature Convolution for Representation Learning

Abstract

Graph convolutional network (GCN) is an emerging neural network approach. It learns new representation of a node by aggregating feature vectors of all neighbors in the aggregation process without considering whether the neighbors or features are useful or not. Recent methods have improved solutions by sampling a fixed size set of neighbors, or assigning different weights to different neighbors in the aggregation process, but features within a feature vector are still treated equally in the aggregation process. In this paper, we introduce a new convolution operation on regular size feature maps constructed from features of a fixed node bandwidth via sampling to get the first-level node representation, which is then passed to a standard GCN to learn the secondlevel node representation. Experiments show that our method outperforms competing methods in semi-supervised node classification tasks. Furthermore, our method opens new doors for exploring new GCN architectures, particularly deeper GCN models.

MeronymNet: A Hierarchical Approach for Unified and Controllable Multi-Category Object Generation

“…To meet this requirement, we model the object layout as an undirected graph. We represent this graph in terms of Feature Matrix 𝑋 and Adjacency Matrix 𝐴 [20,45]. Let p be the maximum possible number of parts across all object categories, i.e.…”

Zero-Shot Multi-View Indoor Localization via Graph Location Networks

“…citation networks [38], WebKB graphs [1]) that have thousands of nodes and edges. Moreover, it was observed in [28,53] that the common instantiation of the a ention mechanism on GNNs (i.e. GATs) does not necessarily bring performance boost over standard GCNs on distinct graph datasets.…”

Rethinking Kernel Methods for Node Representation Learning on Graphs

“…They were introduced in [21], which consist of an iterative process aggregating and transforming representation vectors of its neighboring nodes to capture structural information. Recently, several variants have been proposed, which employ self-attention mechanism [40] or improve network architectures [43,48] to boost the performance. However, most of them are based on empirical intuition and heuristics.…”

“…Therefore, learning to understand graphs is a crucial problem in machine learning. Previous studies in the literature generally fall into two main categories: (1) graph classification [8,21,42,49,50], where the whole structure of graphs is captured for similarity comparison; (2) node classification [1,21,40,43,48], where the structural identity of nodes is determined for representation learning.…”

“…Recent efforts in Graph Convolutional Networks (GCNs) [21] have made great progress on node classification. In particular, these efforts broadly follow a recursive neighborhood aggregation scheme to capture structural information, where each node aggregates feature vectors of its neighbors to compute its new features [1,43,48]. Empirically, these GCNs have achieved the state-of-the-art performance on node classification.…”