Questions tagged [graph-neural-network]
The graph-neural-network tag has no usage guidance.
58
questions
1
vote
1
answer
42
views
What're the standard ways of padding data for GNNs?
I am working on materials property prediction using GNNs with torch_geometric.
Each data in my dataset has different number of feature vectors x, edge_index vectors ...
- 11
0
votes
0
answers
9
views
What is the prior mu in Heterogeneous Graph Transformer?
I am reading https://arxiv.org/pdf/2003.01332.pdf and do not understand what the prior (\mu) is supposed to be. I also found their implementation on github, but it is still not clear to me. For ...
- 101
0
votes
0
answers
96
views
Convert specific domain knowledge text to a knowledge graph
As part of this semester assignment , I'm working on a project that aims to to represent the knowledge in "PMBOK 6th edition, section 11: Project Risk Management (page 395 -> 458)" and the knowledge ...
0
votes
0
answers
8
views
how to derive this negative sampling approximation for graphs?
I am watching CS224W video (https://www.youtube.com/watch?v=Xv0wRy66Big&list=PLoROMvodv4rPLKxIpqhjhPgdQy7imNkDn&index=8) and have some issue understanding the following approximation:
Does ...
- 161
0
votes
0
answers
15
views
Open question - approach to node classification in a directed multi-relational graph
The problem in hand is: In a strategy game, I need to select characters to form my team, in order to win against a second team, chosen by an opponent player. The universe of characters contains U ...
- 101
0
votes
0
answers
25
views
Train test split in Graph Convolution Network for image classification
I am trying to construct a GCN for image classification where each pixel is a single node in the graph. However I want to train and test the model within the same graph, so I constructed a single ...
- 1
0
votes
0
answers
21
views
How to Use Graph Learning Libraries to Predict Edges on a Graph where Each Node Has an Embedding?
An undirectional graph $\mathcal{G}$ has the set of nodes $\mathcal{N}$ where each node has an associated unique embedding of $512$ dimensions. Note that the embeddings themselves are fixed, and not ...
- 315
0
votes
1
answer
244
views
Implementation of Graph Neural Network for Image Classification
I'm currently working on a project where I want to utilize Graph Neural Networks (GNNs) for image classification tasks. However, I'm facing difficulties in understanding how to implement GNNs ...
- 172
0
votes
0
answers
24
views
Fine-tuning Pretrained Models for Web DOM Interaction Prediction Task
I am currently working on a side project that involves predicting changes to a webpage's DOM based on user interactions. The idea is to input the initial DOM state and a user interaction, and predict ...
0
votes
0
answers
24
views
Imbalanced classes
I am currently dealing with imbalanced classes in my case of binary classification, where one class represents only 4% of the other class. To address this issue, here is the approach I have taken:
...
- 1
1
vote
1
answer
58
views
Trouble Training GNN for Binary Node Classification Task
I am using a GNN to solve a problem in which I have a query target and an undirected graph. My goal is to emit a subset of nodes in the graph (via a node-wise binary prediction) whose features sum to ...
- 11
0
votes
0
answers
21
views
Questions regarding Relational Graph Convolutional Network for a Fraud Detection problem
I am currently working on a transaction dataset (https://github.com/IBM/TabFormer/tree/main/data/credit_card) and I intend to build a fraud detection engine, but with tabular data transformed into a ...
0
votes
1
answer
25
views
What are effective ways to merge different parts of a document to form a single document?
I have a dataset where I have 3 different columns (Title, Keyword and Abstract) representing a document. I have to build a text classification model using TextGCN, where documents and words will be ...
0
votes
0
answers
65
views
How to correctly implement the NNConv in Pytorch?
I tried to make a GNN class which can make use of my node features along with my edge features of the graph.I have implemented NNConv in order to use the edge features, but I am not able to understand ...
- 103
0
votes
0
answers
12
views
Terminology question - closed-form QA graph networks
I'm trying to find graph network research on closed-form sequential question answering, but whenever I include 'question-answering' I of course see cutting edge research into free-form conversational, ...
- 111
0
votes
0
answers
19
views
Is there a Graph Neural Network that learns from its neighboring labels (and features)?
I built a heterogeneous graph on a citation graph with a Heterogeneous Graph Convolutional Neural Network in PyTorch and DGL. The graph structure looks like this: (author, writes, paper), (paper, ...
- 1
0
votes
0
answers
50
views
2
votes
1
answer
177
views
Trying to average node values over local neighborhoods in a graph using a GCN
I'm new to Graph Convolutional Networks (and pytorch in general) so I'm trying to verify that the message passing layer is working as expected before I go on to adding layers to the network.
But when ...
- 21
0
votes
0
answers
26
views
how to find out input data, its structure and how to achieve them on graph machine learning model?
I'm a newbie in graph machine learning and apologize if my question is silly. There is a model suggested in some paper for inductive link prediction, I need to use that model on my custom graph but I ...
- 1
0
votes
0
answers
45
views
SGD for Graph Neural Networks
I was going through some research papers about Graph Neural Networks; what struck me is that very often SGD is used as optimiser (as in PointGNN, DGCNN and Graphsage).
I figured that for "regular&...
- 11
0
votes
1
answer
21
views
Is there a procedure for determining if a classification problem is ill-defined?
Consider a group of objects denoted $O = \{o_0, o_1, \cdots\}$ where each object is associated with a feature vector $F = \{f_0, f_1, \cdots\, f_{N-1}\}$. For this case, assume the features are ...
- 123
1
vote
2
answers
61
views
How to cluster components of a graph containing text data?
Suppose that I have a graph that has components like the image below.
Graph nodes contain text data (titles) and the edges data is the similarity (percentage).
I know that each component represents a ...
1
vote
0
answers
23
views
What is purpose of diffusion matrix? What is its role in graph clustering? How do it serve its purpose by equation?
As titled, I am not quite sure what is purpose of Diffusion Matrix, which is support to clustering graph.
From this article (https://arxiv.org/pdf/2006.05582v1.pdf) I am might able to compute the ...
- 11
1
vote
1
answer
249
views
Captum vs GNNExplainer for explainability in Graph Neural Networks
I'm new to Graph Neural Networks and interested in exploring frameworks that allow the identification of nodes/edges that underlie prediction.
I came across :
(1) a model architecture (GNNExplainer) ...
- 111
0
votes
1
answer
34
views
Normalizing softmax by dividing by its maximum?
Reading this paper, I'm struggling to understand the step with the question mark (page 3). The formula for $\textbf r$ uses $\textbf q_i$ (no tilde), but the numeric values in the following paragraph ...
- 103
1
vote
0
answers
69
views
GNN Model - Analyzing Training Curve
Introduction. Actually, I am working on a Graph Neural Network (GNN) model to predict some graph-level float values. So, input=graph, output=float predicted value. I trained and evaluated the proposed ...
- 11
0
votes
0
answers
91
views
How to evaluate Light Graph Convolutional Networks (LightGCN) correctly on sparse binary data?
I implemented the LightGCN at work to recommend k items to users according to the TensorFlow implementation of Microsoft: https://github.com/microsoft/recommenders/blob/main/examples/...
- 109
1
vote
0
answers
12
views
How to convert ECG Data to Graphical Data so that it can be used in GNNs?
I am trying to predict arrythmia using GCNN but the problem i am facing is that the data is in tabular format screenshot attached below. Upon reading i found out that there needs to nodes and edges ...
- 11
0
votes
1
answer
80
views
Incorporating structural information in a Transformer?
For a Neural Machine Translation (NMT) task, my input data has relational information. This relation could be modelled using a graphical structure.
So one approach could be to use Graph Neural Network ...
- 125
3
votes
4
answers
1k
views
How to perform inductive train/test split for GraphSAGE classification
Let's say I have a network that consists of a single weakly connected component. From various papers I've seen that if you want to use inductive GNNs like GraphSAGE, it is advisable to split your ...
0
votes
0
answers
110
views
Why is "Hidden State" at Node in Graph Neural Network Considered "Compressed" Representation?
I am reading article on Graph Neural Network (GNN) and it is mentioned:
The memory stores the states of all the nodes, acting as a compressed representation of the node’s past interactions.
Why is ...
- 155
1
vote
0
answers
35
views
How to define a graph in GNN? [closed]
I am new to graph neural network (GNN). Without knowing a graph in advance, how can we possibly form an adjacency matrix? Assume there are 3 nodes (vertices): A, B, & C. There are could be many ...
- 126
1
vote
0
answers
27
views
Are GNNs/GCNs viable for graphs with no node features, with only the unique node IDs? Are they different from DeepWalk at that point?
I started to dig into GNNs for the first time and I have trouble understanding its advantages over NLP inspired embedding methods like DeepWalk and node2vec. Do GNNs only shine with node features? Or ...
- 11
1
vote
0
answers
35
views
AI algorithm model that outputs a list of unknown length [closed]
I have a dataset with the following x columns:
date
time
is_weekend
is_holiday
start_intersection
end_intersection
The output is a list of intersections, that connect start_intersection with ...
1
vote
0
answers
19
views
Node embeddings and densely connected graphs
I have a dataset of users and their interests (represented by categories) and I’m trying to embed the graph which results from connecting users if they have a common interest, so I’ll add an edge for ...
- 11
1
vote
1
answer
2k
views
How train - test split works for Graph Neural Networks
I have recently started studying GNN's. I have covered GCN and GraphSage so far. But I am confused regarding the process when testing occurs.
Now suppose in the graph above I am using the nodes as ...
- 113
1
vote
1
answer
39
views
How to define similarity between nodes in original graph?
While there has been a lot of talk in how to define the similarity between nodes in the embedding space, but I don't seem to come across any talking about defining the similarity between nodes in the ...
- 411
1
vote
0
answers
53
views
How to create a graph network (Networkx) from the solution of Ordinary differential equations?
I have a list of Nodes.
Suppose,
N =["1","2","3","4"] # there can be different number of nodes
And I have solved the ...
- 103
1
vote
2
answers
595
views
Graph Neural Network fails at generalizing on unseen graph topologies
I'm using PytorchGeometric to train a graph convolutional network for regression over nodes problem (the graph models physical phenomena in the network of sensors; the network of sensors is actually ...
- 11
1
vote
0
answers
14
views
How to construct a graph using Neural Structured Learning framework?
My dataset (both features and label) consists of continuous variables. Dimension of features is (12,). The number of samples are of 7th order of magnitude (about 11 ...
- 111
1
vote
0
answers
151
views
Applying network diffusion to a graph created with the igraph library in R
I have created a graph using the igraph library.
...
0
votes
0
answers
85
views
Validation accuracy does not increase for binary classification using GNN
I am trying to perform graph classification with GINConv model of GNN. I have tried everything from varying dropouts to weight decay (for L2 regularization), learning rate (1e-6 to 1e-3), batch ...
-1
votes
2
answers
35
views
Optimisation of neural networks
Do neural networks get optimized by trial and error, by data scientists, or is there some way of optimizing values through accurate mathematical equations?
4
votes
1
answer
5k
views
What are latent representations?
I am reading some research papers about graph convolutional neural networks and I have seen the term "latent representations" used a lot. For instance, "the model was able to learn ...
- 41
0
votes
1
answer
85
views
Understanding Node Embeddings
I have only just started to look into graph neural networks and I am a little confused on the node embedding process. Here is my understanding, please let me know if i misunderstood:
Given unlabelled ...
- 111
0
votes
2
answers
30
views
Spectral Networks and Deep Locally Connected Networks on Graphs
I’m reading the paper Spectral Networks and Deep Locally Connected Networks on Graphs and I’m having a hard time understanding the notation shown in the picture below (the scribbles are mine):
...
1
vote
0
answers
152
views
Dying gradient issue in Graph Neural Networks
I am using Pytorch-Geometric library to implement a Graph Convolutional Layer(GCN) followed by few linear layers for a prediction task. But after training on graphs with np. of nodes being 10K and no. ...
- 11
1
vote
1
answer
140
views
What is a good approach for embedding both textual and spatial features for document classification?
I am working on a document classifier that can perform the classification based on the document structure as well. My plan is to get the word embedding as well as the word coordinates and somehow ...
- 111
1
vote
0
answers
35
views
(Graph Convolutional Network (GCN) based recommender system maintenance issue [closed]
I have built an item-item recommender model using (Graph Convolutional Network (GCN) for an E-commerce website. Could you please help me with the maintenance of the model.
How often should I retrain ...
- 21
0
votes
0
answers
164
views
Unstable Training when combining Graph Neural Networks for Graph Classification Tasks
I have been combining Graph Convolutional Layers and Graph Pooling layers to define a neural network architecture for Graph Classification tasks. Specifically, using the Graph Convolutional Layer ...
- 41