Sparse training of discrete diffusion models for graph generation
Generative models for graphs often encounter scalability challenges due to the inherent
need to predict interactions for every node pair. Despite the sparsity often exhibited by real …
need to predict interactions for every node pair. Despite the sparsity often exhibited by real …
Optimizing ood detection in molecular graphs: A novel approach with diffusion models
Despite the recent progress of molecular representation learning, its effectiveness is
assumed on the close-world assumptions that training and testing graphs are from identical …
assumed on the close-world assumptions that training and testing graphs are from identical …
Leveraging Graph Diffusion Models for Network Refinement Tasks
Most real-world networks are noisy and incomplete samples from an unknown target
distribution. Refining them by correcting corruptions or inferring unobserved regions typically …
distribution. Refining them by correcting corruptions or inferring unobserved regions typically …
Latent Graph Diffusion: A Unified Framework for Generation and Prediction on Graphs
In this paper, we propose the first framework that enables solving graph learning tasks of all
levels (node, edge and graph) and all types (generation, regression and classification) with …
levels (node, edge and graph) and all types (generation, regression and classification) with …
Editing Partially Observable Networks via Graph Diffusion Models
Most real-world networks are noisy and incomplete samples from an unknown target
distribution. Refining them by correcting corruptions or inferring unobserved regions typically …
distribution. Refining them by correcting corruptions or inferring unobserved regions typically …
Sparse Training of Discrete Diffusion Models for Graph Generation
QIN Yiming, C Vignac, P Frossard - openreview.net
Generative models for graphs often encounter scalability challenges due to the inherent
need to predict interactions for every node pair. Despite the sparsity often exhibited by real …
need to predict interactions for every node pair. Despite the sparsity often exhibited by real …
[PDF][PDF] OMI Research Newsletter–August 2023
M Hoglund, E FERRUCCI, C HERNÁNDEZ - oxford-man.ox.ac.uk
News. The Conference on Learning in Games and Algorithmic Collusion Workshop will take
place on 12 and 13 October 2023 at the University of Oxford. The conference aims at …
place on 12 and 13 October 2023 at the University of Oxford. The conference aims at …