On Representation Knowledge Distillation for Graph Neural Networks

Autor:	Chaitanya K. Joshi, Fayao Liu, Xu Xun, Jie Lin, Chuan Sheng Foo
Rok vydání:	2022
Předmět:	FOS: Computer and information sciences Computer Science - Machine Learning Statistics - Machine Learning Artificial Intelligence Computer Networks and Communications Machine Learning (stat.ML) Software Machine Learning (cs.LG) Computer Science Applications
Zdroj:	IEEE Transactions on Neural Networks and Learning Systems. :1-12
ISSN:	2162-2388 2162-237X
DOI:	10.1109/tnnls.2022.3223018
Popis:	Knowledge distillation is a learning paradigm for boosting resource-efficient graph neural networks (GNNs) using more expressive yet cumbersome teacher models. Past work on distillation for GNNs proposed the Local Structure Preserving loss (LSP), which matches local structural relationships defined over edges across the student and teacher's node embeddings. This paper studies whether preserving the global topology of how the teacher embeds graph data can be a more effective distillation objective for GNNs, as real-world graphs often contain latent interactions and noisy edges. We propose Graph Contrastive Representation Distillation (G-CRD), which uses contrastive learning to implicitly preserve global topology by aligning the student node embeddings to those of the teacher in a shared representation space. Additionally, we introduce an expanded set of benchmarks on large-scale real-world datasets where the performance gap between teacher and student GNNs is non-negligible. Experiments across 4 datasets and 14 heterogeneous GNN architectures show that G-CRD consistently boosts the performance and robustness of lightweight GNNs, outperforming LSP (and a global structure preserving variant of LSP) as well as baselines from 2D computer vision. An analysis of the representational similarity among teacher and student embedding spaces reveals that G-CRD balances preserving local and global relationships, while structure preserving approaches are best at preserving one or the other. Our code is available at https://github.com/chaitjo/efficient-gnns Comment: IEEE Transactions on Neural Networks and Learning Representation (TNNLS), Special Issue on Deep Neural Networks for Graphs: Theory, Models, Algorithms and Applications
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a33e7911455786fd9a42f689917513d5 https://doi.org/10.1109/tnnls.2022.3223018 Zobrazit plný text záznamu