Optimization of graph neural networks: Implicit acceleration by skip connections and more depth

K Xu, M Zhang, S Jegelka… - … on Machine Learning, 2021 - proceedings.mlr.press
Abstract Graph Neural Networks (GNNs) have been studied through the lens of expressive
power and generalization. However, their optimization properties are less well understood …
被引用次数：80 相关文章

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu, M Zhang, S Jegelka, K Kawaguchi - icml.cc
ICML21 copy Page 1 Optimization of Graph Neural Networks: Implicit Acceleration by Skip
Connections and More Depth Keyulu Xu, Mozhi Zhang, Stefanie Jegelka, Kenji Kawaguchi …
[PDF] mlr.press

[PDF][PDF] Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu, M Zhang, S Jegelka, K Kawaguchi - Training - proceedings.mlr.press
Abstract Graph Neural Networks (GNNs) have been studied through the lens of expressive
power and generalization. However, their optimization properties are less well understood …

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu, M Zhang, S Jegelka, K Kawaguchi - arXiv e-prints, 2021 - ui.adsabs.harvard.edu
Abstract Graph Neural Networks (GNNs) have been studied through the lens of expressive
power and generalization. However, their optimization properties are less well understood …

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu - people.csail.mit.edu
Keyulu Xu MIT Page 1 Optimization of Graph Neural Networks: Implicit Acceleration by Skip
Connections and More Depth Keyulu Xu MIT With Mozhi Zhang, Stefanie Jegelka, Kenji …

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu, M Zhang, S Jegelka, K Kawaguchi - arXiv preprint arXiv:2105.04550, 2021 - arxiv.org
Graph Neural Networks (GNNs) have been studied through the lens of expressive power
and generalization. However, their optimization properties are less well understood. We take …

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu, M Zhang, S Jegelka, K Kawaguchi - International Conference on …, 2021 - hal.science
Graph Neural Networks (GNNs) have been studied through the lens of expressive power
and generalization. However, their optimization properties are less well understood. We take …

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu, M Zhang, S Jegelka, K Kawaguchi - International Conference on …, 2021 - par.nsf.gov
Graph Neural Networks (GNNs) have been studied through the lens of expressive power
and generalization. However, their optimization proper-ties are less well understood. We …

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth

K Xu, M Zhang, S Jegelka, K Kawaguchi - Proceedings of the 38th …, 2021 - par.nsf.gov
Graph Neural Networks (GNNs) have been studied through the lens of expressive power
and generalization. However, their optimization properties are less well understood. We take …