Rethinking Graph Neural Architecture Search from Message-passing

By Shaofei Cai, Liang Li, Jincan Deng, Zhengjun Zha, Li Su, Qingming Huang. 2021

64 citations recorded by the source corpus.

Graph neural networks (GNNs) emerged recently as a standard toolkit for learning from data on graphs. Current GNN designing works depend on immense human expertise to explore different message-passing mechanisms, and require manual enumeration to determine the proper message-passing depth. Inspired by the strong searching capability of neural architecture search (NAS) in CNN, this paper proposes Graph Neural Architecture Search (GNAS) with novel-designed search space. The GNAS can automatically learn better architecture with the optimal depth of message passing on the graph. Specifically, we design Graph Neural Architecture Paradigm (GAP) with tree-topology computation procedure and two types of fine-grained atomic operations (feature filtering & neighbor aggregation) from message-passing mechanism to construct powerful graph network search space. Feature filtering performs adaptive feature selection, and neighbor aggregation captures structural information and calculates neighbors’ statistics. Experiments show that our GNAS can search for better GNNs with multiple message-passing mechanisms and optimal message-passing depth. The searched network achieves remarkable improvement over state-of-the-art manual designed and search-based GNNs on five large-scale datasets at three classical graph tasks. Codes can be found at https://github.com/phython96/GNAS-MP.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Materials Chemistry

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xia Hu
2. Qingquan Song
3. Xiao Huang
4. Kaixiong Zhou
5. Quanming Yao
6. Irwin King
7. Fan Yang
8. Ninghao Liu

In Computer Vision and Pattern Recognition

1. Xia Hu
2. Qingquan Song
3. Kaixiong Zhou
4. Xiao Huang

Citations in the corpus

• Auto-GNN: Neural architecture search of graph neural networks (2019, methodology): “For example, targeting at the graph classification problem, previous work (Cai et al., 2021; Wei et al., 2021) incorporates the operation of feature filtration or graph pooling into the search space.”