LookHops: light multi-order convolution and pooling for graph classification
This work provides a more efficient and performant method for graph classification, which is beneficial for researchers and practitioners working with graph-structured data.
This paper addresses the computational cost of higher-order graph convolution and pooling for graph classification. They propose a light k-order method that reduces parameters while improving performance, outperforming SOTA algorithms in efficiency and performance on six graph classification benchmarks.
Convolution and pooling are the key operations to learn hierarchical representation for graph classification, where more expressive $k$-order($k>1$) method requires more computation cost, limiting the further applications. In this paper, we investigate the strategy of selecting $k$ via neighborhood information gain and propose light $k$-order convolution and pooling requiring fewer parameters while improving the performance. Comprehensive and fair experiments through six graph classification benchmarks show: 1) the performance improvement is consistent to the $k$-order information gain. 2) the proposed convolution requires fewer parameters while providing competitive results. 3) the proposed pooling outperforms SOTA algorithms in terms of efficiency and performance.