Unsupervised Multiple Network Alignment with Multinominal GAN and Variational Inference

Zhou, Yang; Ren, Jiaxiang; Jin, Ruoming; Zhang, Zijie; Dou, Dejing; Yan, Da

doi:10.1109/bigdata50022.2020.9377929

Cited by 7 publications

(1 citation statement)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To reduce the time complexity, a line of work leverages the hierarchy of graphs by finding clusters in graphs (Zhang et al 2019;Jing et al 2022b;Wang, Dou, and Zhang 2022) to accelerate the pairwise alignment following a coarsenalign-interpolate strategy. For multi-network alignment, the transitivity constraint was first proposed to ensure the consistency between alignments for different network pairs (Chu et al 2019;Zhang and Philip 2015;Zhou et al 2020), but is hard to handle due to the non-convexity. The product graph, whose size becomes intractable in the multi-network setting, is adopted to model high-order alignment consistency (Du, Liu, and Tong 2021;Li et al 2021).…”

Section: Related Workmentioning

confidence: 99%

Hierarchical Multi-Marginal Optimal Transport for Network Alignment

Zeng,

Du,

Zhang

et al. 2024

AAAI

View full text Add to dashboard Cite

Finding node correspondence across networks, namely multi-network alignment, is an essential prerequisite for joint learning on multiple networks. Despite great success in aligning networks in pairs, the literature on multi-network alignment is sparse due to the exponentially growing solution space and lack of high-order discrepancy measures. To fill this gap, we propose a hierarchical multi-marginal optimal transport framework named HOT for multi-network alignment. To handle the large solution space, multiple networks are decomposed into smaller aligned clusters via the fused Gromov-Wasserstein (FGW) barycenter. To depict high-order relationships across multiple networks, the FGW distance is generalized to the multi-marginal setting, based on which networks can be aligned jointly. A fast proximal point method is further developed with guaranteed convergence to a local optimum. Extensive experiments and analysis show that our proposed HOT achieves significant improvements over the state-of-the-art in both effectiveness and scalability.

show abstract