ML Anthology

Graph Space Embedding

João P. B. Pereira, Albert K. Groen, Erik S. G. Stroes, Evgeni Levin
IJCAI 2019 pp. 3253-3259
doi:10.24963/IJCAI.2019/451 /ijcai/2019/pereira2019ijcai-graph/

Abstract

We propose the Graph Space Embedding (GSE), a technique that maps the input into a space where interactions are implicitly encoded, with little computations required. We provide theoretical results on an optimal regime for the GSE, namely a feasibility region for its parameters, and demonstrate the experimental relevance of our findings. Next, we introduce a strategy to gain insight on which interactions are responsible for the certain predictions, paving the way for a far more transparent model. In an empirical evaluation on a real-world clinical cohort containing patients with suspected coronary artery disease, the GSE achieves far better performance than traditional algorithms.

PDF IJCAI Semantic Scholar

Cite

Text

Pereira et al. "Graph Space Embedding." International Joint Conference on Artificial Intelligence, 2019. doi:10.24963/IJCAI.2019/451

Markdown

[Pereira et al. "Graph Space Embedding." International Joint Conference on Artificial Intelligence, 2019.](https://mlanthology.org/ijcai/2019/pereira2019ijcai-graph/) doi:10.24963/IJCAI.2019/451

BibTeX

@inproceedings{pereira2019ijcai-graph,
  title     = {{Graph Space Embedding}},
  author    = {Pereira, João P. B. and Groen, Albert K. and Stroes, Erik S. G. and Levin, Evgeni},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2019},
  pages     = {3253-3259},
  doi       = {10.24963/IJCAI.2019/451},
  url       = {https://mlanthology.org/ijcai/2019/pereira2019ijcai-graph/}
}