Any2Graph: Deep End-To-End Supervised Graph Prediction With An Optimal Transport Loss
This work addresses the challenge of supervised graph prediction for applications such as satellite imagery analysis and molecular design, offering a versatile solution with broad applicability.
The authors tackled the problem of predicting entire graphs from arbitrary inputs by introducing Any2Graph, a deep learning framework that uses a novel Optimal Transport loss, achieving superior performance on both synthetic and real-world tasks like map construction and molecule prediction.
We propose Any2graph, a generic framework for end-to-end Supervised Graph Prediction (SGP) i.e. a deep learning model that predicts an entire graph for any kind of input. The framework is built on a novel Optimal Transport loss, the Partially-Masked Fused Gromov-Wasserstein, that exhibits all necessary properties (permutation invariance, differentiability and scalability) and is designed to handle any-sized graphs. Numerical experiments showcase the versatility of the approach that outperform existing competitors on a novel challenging synthetic dataset and a variety of real-world tasks such as map construction from satellite image (Sat2Graph) or molecule prediction from fingerprint (Fingerprint2Graph).