Dual GNNs: Learning Graph Neural Networks with Limited Supervision
2. prosince 2022
Řečníci
Abdullah Alchihabi
Řečník · 0 sledujících
Yuhong Guo
Řečník · 0 sledujících
O prezentaci
Graph Neural Networks (GNNs) require a relatively large number of labeled nodes and a reliable/uncorrupted graph connectivity structure in order to obtain good performance on the semi-supervised node classification task. The performance of GNNs can degrade significantly as the number of labeled nodes decreases or the graph connectivity structure is corrupted by adversarial attacks or due to noises in data measurement/collection. Therefore, it is important to develop GNN models that are able to achieve good performance when there is limited supervision knowledge – a few labeled nodes and noisy graph structures. In this paper, we propose a novel Dual GNN learning framework to address this challenging task. The proposed framework has two GNN based node prediction modules. The primary module uses the input graph structure to induce regular node embeddings and predictions with a regular GNN baseline, while the auxiliary module constructs a new graph structure through fine-grained spectral clustering and learns new node embeddings and predictions. By integrating the two modules in a dual GNN learning framework, we perform joint learning in an end-to-end fashion. This general framework can be applied on many GNN baseline models. The experimental results validate that the proposed dual GNN framework can greatly outperform the GNN baseline methods when the labeled nodes are scarce, and the graph connectivity structure is noisy.Graph Neural Networks (GNNs) require a relatively large number of labeled nodes and a reliable/uncorrupted graph connectivity structure in order to obtain good performance on the semi-supervised node classification task. The performance of GNNs can degrade significantly as the number of labeled nodes decreases or the graph connectivity structure is corrupted by adversarial attacks or due to noises in data measurement/collection. Therefore, it is important to develop GNN models that are able to a…
Organizátor
NeurIPS 2022
Účet · 961 sledujících
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
How to Handle Distributional Shifts? Challenges, Research Progress and Path Forward
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
LAMP: Extracting Text from Gradients with Language Model Priors
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 1 diváků, což je 0.1 %
Expansion and Shrinkage of Localization for Weakly-Supervised Semantic Segmentation
Zhlédnout později
Jinlong Liu, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Closing remarks
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Understanding stock market instability via graph auto-encoders
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Take 5: Interpretable Image Classification with a Handful of Features
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %