Questions tagged [graph-neural-networks]
For questions related to graph neural networks, which are artificial neural networks applied to graphs.
85
questions
3
votes
2
answers
351
views
How can we derive a Convolution Neural Network from a more generic Graph Neural Network?
Convolution Neural Network (CNNs) operate over strict grid-like structures ($M \times N \times C$ images), whereas Graph Neural Networks (GNNs) can operate over all-flexible graphs, with an undefined ...
CommunityBot
- 1
0
votes
0
answers
49
views
ST-GCN: graph convolution operator in Geometry-Aware Interaction Network (GAIN)
I need help implementing the model in this paper:
They have adopted spatio-temporal graph convolution operator in ST-GCN [section 3.1.2]. I've found there is popular libraries available for GCN: ...
0
votes
0
answers
18
views
Link Prediction Task on Multigraphs with GraphSAGE
I need to perform a link prediction task on a heteronegous multigraph (multi-node types, multi-edge types, multi-edges between pairs of nodes, node features and edge features) in the inductive setting....
- 1
-1
votes
1
answer
106
views
What is GNN Cheatsheet in PyG Docs [closed]
I am going through the Pytorch Geometric documentation: https://pytorch-geometric.readthedocs.io/en/latest/index.html which is built on Pytorch .Here they mentioned about GNN Cheatsheet: https://...
1
vote
1
answer
674
views
Are there Explainable GNN methods for node regression tasks?
I am wondering if there are gnn explainable methods for a regression task (e.g., traffic forecasting) where nodes have numerical features and the predicted output is a numerical value. Most of ...
CommunityBot
- 1
1
vote
2
answers
43
views
How to do image classification with optional metadata?
I have a vanilla image classification problem. The image may optionally have some numerical metadata associated with it. We don't assume uniform availability of this metadata, i.e., the model should ...
- 206
1
vote
1
answer
81
views
Why is there a shared matrix W in graph attention networks instead of the query-key-value trio like in regular transformers?
In section 2.1 of the Graph attention network paper
The graph attention layer is described as
as an initial step, a shared
linear transformation, parametrized by a weight matrix, W ∈ RF ′×F , is ...
- 483
0
votes
1
answer
132
views
Model Suggestion for graph data
I am trying to figure out the right model/algorithm for a graph dataset to develop a machine learning pipeline. I have looked into Graph Neural Network(GNN) but all of the tutorials I found, trained ...
CommunityBot
- 1
1
vote
2
answers
1k
views
Can I extend Graph Convolutional Networks to graphs with weighted edges?
I'm researching spatio-temporal forecasting utilising GCN as a side project, and I am wondering if I can extend it by using a graph with weighted edges instead of a simple adjacency matrix with 1's ...
CommunityBot
- 1
0
votes
0
answers
14
views
Requesting resources on causal networks for 2D strategy game
I am requesting research, articles, abstracts or interesting opinions that will help me create a complex causal neural network. There are many detailed resources on causal discovery, image recognition,...
3
votes
1
answer
66
views
How can I improve this toy Graph Neural Network Generative Language model [closed]
Background
I'm an undergraduate student with research interests in a field of physics that has significant overlap with graph theory, and a functioning knowledge of how simple neural nets work and how ...
- 483
0
votes
1
answer
36
views
Do GNNs operate on enitre graphs or do they basically iterate over each node one-by-one?
I understand how GNNs/GCNs aggregate an arbitrary number of nodes' information from the neighborhood of a target node in order to predict an attribute of that target node. What I don't understand is, ...
CommunityBot
- 1
1
vote
1
answer
325
views
How does Weight Sharing with the Generalization in Graph Neural Networks work?
I have two closely related points regarding the weight sharing and generalization of graph Neural network. For illustration purposes, I attached two images which I reference. Images are taken from the ...
CommunityBot
- 1
2
votes
1
answer
915
views
What are examples of node 'features' in graph networks?
Context: I was reading Chapter 3 in the following book (here) about graph representation learning. Before I get to node embeddings, I wanted to make sure that I do understand what is meant by the ...
CommunityBot
- 1
0
votes
1
answer
66
views
Which models can be applied recursively?
I come from a math background, so I am not up-to-date with machine learning literature.
For the purpose of learning dynamics, I would like to train a model to minimize the following loss:
$$\mathcal{L}...
- 406
0
votes
0
answers
25
views
Several kinds of edges in a GNN
I have an existing implementation (written by somebody else) of an MPNN using the graph_nets library. The graph net is based on a tree, but has 4 times as many edges: if U is the parent of V and R is ...
- 101
0
votes
1
answer
102
views
Why is the output of my graph neural network not permutation equivariant?
I am using Pytorch to train a graph neural network on a 4x4 graph. Each node has one feature, and the output has one feature. Essentially, the architecture of my GNN looks like this (I'm training the ...
- 111
2
votes
1
answer
125
views
Node classification with random labels for GNNs
I decided to train GCN on the Cora dataset for the node classification task, however, with the random labels, i.e., applying np.random.shuffle(labels). For the ...
- 1,536
2
votes
1
answer
97
views
Why readout operation in message passing graph neural nets have to be invariant to node permutations?
I am reading the paper Neural Message Passing for Quantum Chemistry by Justin Gilmer et al. And I have a question regarding this passage
The message functions $M_t$, vertex update functions $U_t$, ...
CommunityBot
- 1
0
votes
0
answers
12
views
How can one incorporate spatial correlations into time series forecasting?
I am working on a project, where I am trying to predict temperatures of various streets and I have their locations recorded.
I was wondering if I could somehow train a model that could incorporate ...
1
vote
1
answer
31
views
GREED - preservation theoretical properties in the GED(graph edit distance) pridiction
In this paper "GREED: A Neural Framework for Learning Graph Distance Functions", function F is defined to satisfy metric property and triangle inequality property.
I wonder how can I prove ...
- 1,536
1
vote
0
answers
164
views
How to get ZINC 500k dataset?
I have been using the ZINC graph regression dataset available through pytorch geometric datasets for a while now in two of its modes (12k examples and 250k examples). However, in the PapersWithCode ...
- 201
4
votes
1
answer
250
views
Relevance of Weisfeiler–Lehman Graph Isomorphism Test limitation for Graph Neural Networks
Graph Neural Networks power is limited by the power of Weisfeiler–Lehman Graph Isomorphism algorithm.
Quoting wikipedia:
It has been demonstrated that GNNs cannot be more expressive than the
...
- 1,536
1
vote
0
answers
23
views
Is there another type of NN that can capture just the structure of the graph?
I have a classification problem where the inputs are graphs, with no special features in the nodes of the graph.
I tried to use message passing layers like GCN and GIN but they were not able to ...
- 39.7k
1
vote
1
answer
110
views
Are there any advantages of encoding an image as a graph to use in Graph Convolutional Networks?
I have seen this encoding of an image as a graph:
The set of the nodes $V$ is the set of pixels. If the image is of size $10\times10$, then we have $10\cdot10=100$ pixels.
Each node has a length 3 ...
- 1,536
0
votes
0
answers
23
views
Is there a test to determine if a feature space is sufficient for a classification problem?
I am working with GNNs for a node classification problem, and I am only able to achieve about 50% accuracy for the training set. I am not able to overfit the model to the training data. This makes me ...
CommunityBot
- 1
0
votes
1
answer
33
views
How to reduce variance in F1 scores of GAT across multiple runs while using PU Loss?
I am training GAT using a custom loss function(PU Loss) on the Cora and Citeseer dataset. My training file looks like
...
- 35
0
votes
0
answers
61
views
Understanding relation between parameter sharing and Message Passing in Graph Neural Networks
Ravanbakhsh has clearly stated the relation between equivariance and parameter-sharing in neural networks. What I'm missing though, is where (and how) this relation becomes clear by considering the ...
- 307
1
vote
1
answer
111
views
Is there a neural network method to encode a directed graph?
I want to do a graph classification task. Those graphs are directed, and their edges have features. I knew little about graph representation methods, but I did some research, and find most of the ...
- 1,536
0
votes
1
answer
33
views
Is it possible to perform node-level classification and graph-level classification on the same graph? [closed]
I would like to classify the nodes of each graph in a multigraph, and transform the graph structure (or delete some of the nodes).
And then I want to do a graph-level classification problem.
Are there ...
- 5,218
7
votes
3
answers
2k
views
Is there an open-source implementation for graph convolution networks for weighted graphs?
Currently, I'm using a Python library, StellarGraph, to implement GCN. And I now have a situation where I have graphs with weighted edges. Unfortunately, StellarGraph doesn't support those graphs
I'm ...
-1
votes
1
answer
55
views
Can anyone help me how this code extracts features from the graph? [closed]
I have this code from DGCNN Neural Network but i don't understand how it extracts features.
In particular i understand that we get the top knn point but i don't understand the idx_base.
...
- 1,536
1
vote
1
answer
402
views
Rationalle behind SE3 Transformer?
I have just finished reading the SE3 transformer paper (SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks) by Fuchs et-al and although I'm sure I understand less than 50% of the ...
- 121
1
vote
0
answers
90
views
How to learn how to select a subgraph via reinforcement learning?
I have the following problem.
I am given a graph with a lot (>30000) nodes. Nodes are associated with a low (<10)-dimensional feature vector, and edges are associated with a low (<10)-...
- 39.7k
2
votes
1
answer
171
views
Why don't we use diffusion for non-graph CNNs?
I'm pretty new to graph neural networks, so please forgive me if this is a silly question.
Diffusion is a method used to improve graph CNNs, however it seems to me that general CNNs can also benefit ...
- 1,536
1
vote
0
answers
82
views
Temporal Graph Neural Network for motion prediction
Temporal Graph Neural Networks have been used for motion prediction (or traffic forecasting) in the following recent papers:
Dynamic Multiscale Graph Neural Networks for 3D Skeleton-Based Human Motion ...
- 169
1
vote
1
answer
463
views
Using GraphSAGE model for multigraph datasets
I checked out applications of GraphSAGE and it seems like its primarily used for single graph datasets. For example - Cora dataset - Its one big graph with 2708 nodes and 5429 edges. The model can ...
- 1,536
0
votes
1
answer
190
views
How to understand the GCN equation?
I understand GCN does message passing with its neighbours to learn the node embedding.
But I don't understand the following equation.
What "tilda" is referring to equation ...
- 1,751
1
vote
1
answer
563
views
What kind of features does each node have as an input graph to a graph neural network?
What kind of features does each node have as an input graph to a graph neural network? For example, we want to do image classification with GNN, what are the features of each pixel? Or if anyone could ...
CommunityBot
- 1
24
votes
5
answers
20k
views
What is non-Euclidean data?
What is non-Euclidean data?
Here are some sub-questions
Where does this type of data arise? I have come across this term in the context of geometric deep learning and graph neural networks.
...
1
vote
1
answer
183
views
How to use structural information in a Transformer?
I am performing a Neural Machine Translation (NMT) task. In my case, input data has relational information.
I know I can use a Graph Neural Network (GNN) and use a Graph2Seq model. But I can't find a ...
- 206
1
vote
0
answers
24
views
In Graph Neural Network is Message Passing Step Agnostic of Output Values during Training?
So Graph Neural Networks is about representation learning where initially representation of graph is learned in the form of node embeddings. My question is: Are the output values back propagated and ...
- 145
0
votes
1
answer
158
views
What is the best GNN for a NMT task?
I am doing a machine translation task using a Graph2Seq graph neutral network.
There are many different variants of GNN:
GCN
GAT
GraphSage
GGNN
Which one would be the most effective for a machine ...
- 31
1
vote
1
answer
611
views
Is "node embedding" in GNN analogous to "hidden layer" of FFN?
So in Graph Neural Network (GNN) we have node embeddings which is a feature vector that describes the node, is it analogous to hidden layer of Artificial neural network such as feed-forward neural ...
- 39.7k
1
vote
0
answers
27
views
How to model graph node as priority list over a visual scene in neuro-symbolic AI?
Suppose if we have a visual scene graph and we model each component in the scene as a node of a graph and edges which are relationship between the visual scene components. Some of the nodes are like ...
- 31
1
vote
0
answers
78
views
Graph Neural Networks: Why do papers use very low label rates?
I was recently reading the following paper: "Semi-supervised classification with Graph Convolutional Networks" by Kipf and Welling (here).
Question: When testing on datasets, why are the ...
- 165
2
votes
0
answers
63
views
Given a 2-layer GCN, can we choose the dimensions of the 2nd weight matrix, such that this architecture has the same capacity as a 1-layer GCN?
This might be more of a question about nested function classes:
For $k$ class node classification in a graph with $n$ nodes, and $d$ feature vector.
I want to compare
Architecture I: the GCN model of ...
- 39.7k
14
votes
2
answers
10k
views
What is the difference between graph convolution in the spatial vs spectral domain?
I've been reading different papers regarding graph convolution and it seems that they come into two flavors: spatial and spectral. From what I can see the main difference between the two approaches is ...
- 39.7k
2
votes
0
answers
212
views
Are spectral approaches to Graph Neural Networks still considered?
I've been reading several papers and reviews about Graph Neural Networks, and I still feel a bit confused about the difference between the two approaches, and also if the spatial approaches have ...
- 39.7k
0
votes
0
answers
202
views
How does Chebyshev approximation of spectral convolution work?
I was reading the following paper: here. In it, it talks about spectral graph convolutions and says:
We consider spectral convolutions on graphs defined as the
multiplication of a signal $x \in R^N$ (...
- 39.7k