A Framework for Differentiable Supervised Graph Prediction
Abstract
We introduce a general framework to train a deep neural network to output a graph from a variety of input modalities. The framework is built using a novel Optimal Transport loss that exhibits all necessary properties (permutation invariance and differentiability) and allows for handling graphs of any size. We showcase the versatility and state-of-the-art performances of the proposed approach on various real-world tasks and a novel challenging synthetic dataset.
Cite
Text
Krzakala et al. "A Framework for Differentiable Supervised Graph Prediction." ICML 2024 Workshops: Differentiable_Almost_Everything, 2024.Markdown
[Krzakala et al. "A Framework for Differentiable Supervised Graph Prediction." ICML 2024 Workshops: Differentiable_Almost_Everything, 2024.](https://mlanthology.org/icmlw/2024/krzakala2024icmlw-framework/)BibTeX
@inproceedings{krzakala2024icmlw-framework,
title = {{A Framework for Differentiable Supervised Graph Prediction}},
author = {Krzakala, Paul and Yang, Junjie and Flamary, Rémi and d'Alché-Buc, Florence and Laclau, Charlotte and Labeau, Matthieu},
booktitle = {ICML 2024 Workshops: Differentiable_Almost_Everything},
year = {2024},
url = {https://mlanthology.org/icmlw/2024/krzakala2024icmlw-framework/}
}