Graph Neural Networks
-
Upload
khangminh22 -
Category
Documents
-
view
0 -
download
0
Transcript of Graph Neural Networks
Graph
Dongguk AI. LAB. Sung-eun Jang
Definition of Graph
A
B
C
E
DF
β’ Graph
β’ A collection of objects where some pairs of
objects are connected by links
β’ Components of a Graph
β’ Objects nodes, vertices π
β’ Interactions links, edges πΈ
β’ System network, graph πΊ(π, πΈ)
Graph
Dongguk AI. LAB. Sung-eun Jang
Structure of Graph
β’ Directed / Undirected
β’ Weighted / Unweighted
5
5
10
10
11 1
GraphGraph Representation
34
1
2
Adjacency matrix
Node-feature matrix
0 1 0 1
1 0 0 1
0 0 0 1
1 1 1 0
2 0 0 0
0 2 0 0
0 0 3 0
0 0 0 1
2 -1 0 -1
1 2 0 -1
0 0 3 -1
-1 -1 -1 2
Degree matrix Laplacian matrix
π·ππ =
π~π
π΄πππ΄ = ΰ΅π΄ππ = 1 ππ π‘βπππ ππ ππππ
π΄ππ = 0 ππ π‘βπππ ππ ππ ππππ πΏ = π΄ β π·
Dongguk AI. LAB. Sung-eun Jang
GraphWhy Graph?
Dongguk AI. LAB. Sung-eun JangDongguk AI. LAB. Sung-eun Jang
β’ Universal Language for describing complex data
β’ Shared vocabulary between fields
β’ Data availability & computational challenges
Node EmbeddingsDefinition.
Dongguk AI. LAB. Sung-eun Jang
π(π’)
π’
π§π’
β’ Node embedding function (f) map each node in a graph into a low-dimensional space
β’ similarity in the embedding space approximates similarity in the graph
β’ Could be helpful with tasks such asβ¦
β node classification, graph classification, link predictionβ¦
Node EmbeddingsShallow Encoders
Dongguk AI. LAB. Sung-eun Jang
Node π’
Node π£
embedding lookup
nodesimilarity
β’ Use a lookup table consisting of unique embedding of every nodes in graph
β’ Limitations
β’ O(|V|) parameters are needed
β’ Inherently Transductive
β’ Do not incorporate node features
Node EmbeddingsGraph and Modern Deep Learning
Dongguk AI. LAB. Sung-eun Jang
image
textβ’ Arbitrary size
β’ Complex topological structure
β’ No fixed node ordering
β’ Dynamic and multimodal features
graph
Graph Convolutional NetworkGCN
Dongguk AI. LAB. Sung-eun Jang
β’ Transform information at the neighbors and combine it
1) Transform βmessagesβ βπ from neighbors:
2) Add them up
image graph
π0 β0
π
ππ βπ
π1 β1
π2 β2
π3 β3 β¦
Graph Convolutional NetworkCompute Node Features
Dongguk AI. LAB. Sung-eun Jang
β’ Generate node embeddings based on local network neighborhoods
β’ Nodes have embeddings at each layer, repeating combine messagesfrom their neighbor using neural networks
Layer-0Layer-1Layer-2
average messages from neighbors
apply neural network
Graph Convolutional NetworkLimitation of GCN
Dongguk AI. LAB. Sung-eun Jang
?????
β’ For training of the embeddings, model requires that all nodes in the graph are present during training time
β’ As the number of nodes increases, the matrix becomes larger and computational cost increases accordingly.
GraphSAGEDefinition
Dongguk AI. LAB. Sung-eun Jang
β’ GraphSAGE (SAmple and aggreGatE)
β’ Instead of training individual embeddings for each node, generates embeddings by sampling features from neighborhoods
βmini batch
β’ Train a set of aggregator functions that learn to aggregate feature information a nodeβs local neighborhood
β aggregating
GraphSAGEWeighting factor in GraphSAGE
Dongguk AI. LAB. Sung-eun Jang
πΌπ£π’ =1
|π(π£)|
Weighting factor
β’ πΌπ£π’ (importance) is defined explicitly based on the structure properties of graph
β’ All neighbors π’ β π(π£) are equally important to node π£
Graph Attention NetworkIdea
Dongguk AI. LAB. Sung-eun Jang
β’ Specify arbitrary importances to different neighbors of each node in the graph
β’ Compute embedding βπ£π of each node in the graph following an attention strategy:
β Nodes attend over their neighborhoodsβ message
β Implicitly specifying different weights to different nodes
Graph Attention NetworkSelf attention
Dongguk AI. LAB. Sung-eun Jang
πβ π₯
ππ π₯
ππ π₯
πΌππ
πβπ
βπ
Reference
β’ Kipf, T. N., & Welling, M. (2016). Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907.
β’ Hamilton, W., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. In Advances in Neural Information Processing Systems (pp. 1024-1034).
β’ VeliΔkoviΔ, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., & Bengio, Y. (2017). Graph attention networks. arXiv preprint arXiv:1710.10903.
β’ http://web.stanford.edu/class/cs224w/
β’ http://dmqm.korea.ac.kr/activity/seminar/267
β’ https://www.cs.ubc.ca/~lsigal/532S_2018W2/Lecture18a.pdf
Lorem Ipsum is simply dummy text of the printing and typesetting industry.
Dongguk AI. LAB. Sung-eun Jang
μ λͺ©μ μ λ ₯νμΈμLorem Ipsum is simply dummy text of the printing and typesetting industry.
0
1
2
3
4
5
6
νλͺ© 1 νλͺ© 2 νλͺ© 3 νλͺ© 4
μ°¨νΈ μ λͺ©
κ³μ΄ 1 κ³μ΄ 2 κ³μ΄ 3
Graph Attention Network
Dongguk AI. LAB. Sung-eun Jang
1/6
Lorem Ipsum is simply dummy text of the printing and typesetting industry.
μ λͺ©μ μ λ ₯νμΈμLorem Ipsum is simply dummy text of the printing and typesetting industry.
ν맀
1λΆκΈ° 2λΆκΈ° 3λΆκΈ° 4λΆκΈ°
Graph Convolutional Network 1/4
Lorem Ipsum is simply dummy text of the printing and typesetting industry.
Dongguk AI. LAB. Sung-eun Jang