• Publications
  • Influence
How Powerful are Graph Neural Networks?
TLDR
This work characterize the discriminative power of popular GNN variants, such as Graph Convolutional Networks and GraphSAGE, and show that they cannot learn to distinguish certain simple graph structures, and develops a simple architecture that is provably the most expressive among the class of GNNs.
Representation Learning on Graphs with Jumping Knowledge Networks
TLDR
This work explores an architecture -- jumping knowledge (JK) networks -- that flexibly leverages, for each node, different neighborhood ranges to enable better structure-aware representation in graphs.
Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels
TLDR
A new class of graph kernels, Graph Neural Tangent Kernels (GNTKs), which correspond to infinitely wide multi-layer GNNs trained by gradient descent are presented, which enjoy the full expressive power ofGNNs and inherit advantages of GKs.
What Can Neural Networks Reason About?
TLDR
This framework offers an explanation for the empirical success of popular reasoning models, and suggests their limitations, and unify seemingly different reasoning tasks via the lens of a powerful algorithmic paradigm, dynamic programming (DP).
How Neural Networks Extrapolate: From Feedforward to Graph Neural Networks
TLDR
The success of GNNs in extrapolating algorithmic tasks to new data relies on encoding task-specific non-linearities in the architecture or features, and a hypothesis is suggested for which theoretical and empirical evidence is provided.
Are Girls Neko or Shōjo? Cross-Lingual Alignment of Non-Isomorphic Embeddings with Iterative Normalization
TLDR
Iterative Normalization consistently improves word translation accuracy of three CLWE methods, with the largest improvement observed on English-Japanese (from 2% to 44% test accuracy).
Distributional Adversarial Networks
TLDR
Inspired by discrepancy measures and two-sample tests between probability distributions, a framework for adversarial training that relies on a sample rather than a single sample point as the fundamental unit of discrimination is proposed.
GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training
TLDR
A principled normalization method, Graph Normalization (GraphNorm), where the key idea is to normalize the feature values across all nodes for each individual graph with a learnable shift, which improves generalization of GNNs, achieving better performance on graph classification benchmarks.
Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth
TLDR
This work analyzes linearized GNNs and proves that despite the non-convexity of training, convergence to a global minimum at a linear rate is guaranteed under mild assumptions that are validated on real-world graphs.
Generating Random Spanning Trees via Fast Matrix Multiplication
TLDR
The best algorithm for dense graphs can produce a uniformly random spanning tree of an n-vertex graph in time \(O(n^{2.38})\).
...
1
2
...