Graph Classification with 2D Convolutional Neural Networks
This work addresses the problem of efficient and accurate graph classification for researchers and practitioners in machine learning, offering a novel representation approach that simplifies the use of existing CNN architectures.
The paper tackles the challenge of applying convolutional neural networks (CNNs) to graph classification by representing graphs as multi-channel image-like structures, enabling the use of vanilla 2D CNNs. Experiments show that this method outperforms state-of-the-art graph kernels and graph CNNs on 4 out of 6 real-world datasets and offers better time complexity than graph kernels.
Graph learning is currently dominated by graph kernels, which, while powerful, suffer some significant limitations. Convolutional Neural Networks (CNNs) offer a very appealing alternative, but processing graphs with CNNs is not trivial. To address this challenge, many sophisticated extensions of CNNs have recently been introduced. In this paper, we reverse the problem: rather than proposing yet another graph CNN model, we introduce a novel way to represent graphs as multi-channel image-like structures that allows them to be handled by vanilla 2D CNNs. Experiments reveal that our method is more accurate than state-of-the-art graph kernels and graph CNNs on 4 out of 6 real-world datasets (with and without continuous node attributes), and close elsewhere. Our approach is also preferable to graph kernels in terms of time complexity. Code and data are publicly available.