ICML 2021spotlight29 citations

Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approach

Federico Lopez, Beatrice Pozzetti, Steve Trettel, Michael Strube, Anna Wienhard

Abstract

Learning faithful graph representations as sets of vertex embeddings has become a fundamental intermediary step in a wide range of machine learning applications. We propose the systematic use of symmetric spaces in representation learning, a class encompassing many of the previously used embedding targets. This enables us to introduce a new method, the use of Finsler metrics integrated in a Riemannian optimization scheme, that better adapts to dissimilar structures in the graph. We develop a tool to analyze the embeddings and infer structural properties of the data sets. For implementation, we choose Siegel spaces, a versatile family of symmetric spaces. Our approach outperforms competitive baselines for graph reconstruction tasks on various synthetic and real-world datasets. We further demonstrate its applicability on two downstream tasks, recommender systems and node classification.

BibTeX
@InProceedings{pmlr-v139-lopez21a,
  title = 	 {Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approach},
  author =       {Lopez, Federico and Pozzetti, Beatrice and Trettel, Steve and Strube, Michael and Wienhard, Anna},
  booktitle = 	 {Proceedings of the 38th International Conference on Machine Learning},
  pages = 	 {7090--7101},
  year = 	 {2021},
  editor = 	 {Meila, Marina and Zhang, Tong},
  volume = 	 {139},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {18--24 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v139/lopez21a/lopez21a.pdf},
  url = 	 {https://proceedings.mlr.press/v139/lopez21a.html},
  abstract = 	 {Learning faithful graph representations as sets of vertex embeddings has become a fundamental intermediary step in a wide range of machine learning applications. We propose the systematic use of symmetric spaces in representation learning, a class encompassing many of the previously used embedding targets. This enables us to introduce a new method, the use of Finsler metrics integrated in a Riemannian optimization scheme, that better adapts to dissimilar structures in the graph. We develop a tool to analyze the embeddings and infer structural properties of the data sets. For implementation, we choose Siegel spaces, a versatile family of symmetric spaces. Our approach outperforms competitive baselines for graph reconstruction tasks on various synthetic and real-world datasets. We further demonstrate its applicability on two downstream tasks, recommender systems and node classification.}
}
Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approach · ICML 2021