Most graph neural networks (GNNs) use the message passing paradigm, in which node features are propagated on the input graph. Recent works pointed to the distortion of information flowing from distant nodes as a factor limiting the efficiency of message passing for tasks relying on long-distance interactions. This phenomenon, referred to as 'over-squashing', has been heuristically attributed to graph bottlenecks where the number of $k$-hop neighbors grows rapidly with $k$. We provide a precise description of the over-squashing phenomenon in GNNs and analyze how it arises from bottlenecks in the graph. For this purpose, we introduce a new edge-based combinatorial curvature and prove that negatively curved edges are responsible for the over-s...
Graphical models are used to describe the interactions in structures, such as the nodes in decoding ...
Defining the geometry of networks is typically associated with embedding in low-dimensional spaces s...
This paper studies the expressive power of graph neural networks falling within the message-passing ...
Recent works have investigated the role of graph bottlenecks in preventing long-range information pr...
Over-squashing and over-smoothing are two critical issues, that limit the capabilities of graph neur...
International audienceGraph Neural Networks (GNNs) have succeeded in various computer science applic...
Deploying graph neural networks (GNNs) on whole-graph classification or regression tasks is known to...
Training deep graph neural networks (GNNs) poses a challenging task, as the performance of GNNs may ...
Message-passing graph neural networks (MPNNs) emerged as powerful tools for processing graph-structu...
In designing and applying graph neural networks, we often fall into some optimization pitfalls, the ...
International audienceWe analyze graph smoothing with \emph{mean aggregation}, where each node succe...
We analyze graph smoothing with \emph{mean aggregation}, where each node successively receives the a...
Increasing the depth of GCN, which is expected to permit more expressivity, is shown to incur perfor...
Graph Neural Networks (GNNs) have achieved promising performance on a wide range of graph-based task...
Most graph neural networks (GNNs) rely on the message passing paradigm to propagate node features an...
Graphical models are used to describe the interactions in structures, such as the nodes in decoding ...
Defining the geometry of networks is typically associated with embedding in low-dimensional spaces s...
This paper studies the expressive power of graph neural networks falling within the message-passing ...
Recent works have investigated the role of graph bottlenecks in preventing long-range information pr...
Over-squashing and over-smoothing are two critical issues, that limit the capabilities of graph neur...
International audienceGraph Neural Networks (GNNs) have succeeded in various computer science applic...
Deploying graph neural networks (GNNs) on whole-graph classification or regression tasks is known to...
Training deep graph neural networks (GNNs) poses a challenging task, as the performance of GNNs may ...
Message-passing graph neural networks (MPNNs) emerged as powerful tools for processing graph-structu...
In designing and applying graph neural networks, we often fall into some optimization pitfalls, the ...
International audienceWe analyze graph smoothing with \emph{mean aggregation}, where each node succe...
We analyze graph smoothing with \emph{mean aggregation}, where each node successively receives the a...
Increasing the depth of GCN, which is expected to permit more expressivity, is shown to incur perfor...
Graph Neural Networks (GNNs) have achieved promising performance on a wide range of graph-based task...
Most graph neural networks (GNNs) rely on the message passing paradigm to propagate node features an...
Graphical models are used to describe the interactions in structures, such as the nodes in decoding ...
Defining the geometry of networks is typically associated with embedding in low-dimensional spaces s...
This paper studies the expressive power of graph neural networks falling within the message-passing ...