SlideShare a Scribd company logo
RGE: A Repulsive Graph Rectification for Node
Classification via Influence
Jaeyun Song, Sung-Yub Kim, Eunho Yang
ICML 2023
Graduate School of AI, KAIST
Introduction
● Graph Neural Networks (GNNs) are susceptible to structural noises in graphs.
● Chen et al. (2022a) proposed an influence function for individual edges to estimate
the counterfactual effects of removing them.
● We first identify the group influence estimation error of Exhaustive Group
Elimination (EGE), that eliminates all opponent edges at once and propose
Repulsive Group Elimination (RGE), that removes distant edges over neighboring
edges to reduce group effect in GNNs.
A single harmful edge (dotted) can negatively affect the
predictions of multiple nodes (colored).
Problem
• Our goal is to find opponent edges 𝐸′ ⊆ 𝐸 that can improve the test performance when
remove the original graph 𝐺 = (𝑉, 𝐸) . Formally, this can be formulated as
• To avoid the combinatorial explosion of selecting an edge set, we consider a simple
strategy, called Exhaustive edge Group Elimination (EGE):
• The key assumption for the optimality of EGE is the additivity of individual influences.
Graph without 𝐸′
Influence of removing 𝐸′
Problem
• Even though the influence additivity cannot be guaranteed in general graphs, we found
the influence additivity can be established between distant edges for the widely used
GNNs, including GCN, SGC, GAT, and GraphSAGE.
• Proposition 3.1 (Influence additivity for distant edges) Let edges 𝑒, 𝑒′
∈ 𝐸 be distant edges
and, for them, assume that the removal of edge 𝑒 does not change the gradient of
unaffected train node 𝒗 :
for 𝑣 ∈ 𝑉𝑇𝑟 and 𝑣 ∉ 𝑉𝑇𝑟(𝑒). Then, the influence additivity holds.
• Detailed proofs can be found in Appendix B!
Method Overview
• The overview of our method
• Compute the influence for all edges
• Remove non-adjacent edges (distant edges) with negative influence
• Retrain GNNs on graphs without those edges
Repeat these processes until there are no more negative edges
labeled nodes unlabeled nodes non-negative edges negative edges
Remove negative
distant edges
& Retrain GNNs
Repeat processes
Rectified graph
Method
• Repulsive selection rule & Multi-step edge elimination (under 1-layer SGC)
labeled nodes unlabeled nodes non-negative edges negative edges
Choose the most
negative edge
Choose the second
most negative
non-adjacent edge
selected edges
1-st Iteration
Remove edges
& Retrain GNNs
Choose the most
negative edge
2-nd Iteration
Remove edges
& Retrain GNNs
Repeat processes
Experiment: Group Influence Estimation Errors
• RGE significantly decreases the group influence estimation errors
EGE RGE
Group Influence Estimation Errors
Experiment: Performance on Homophilous Graphs
• RGE outperforms other baselines
Experiment: Performance under Other Architectures
• Rectified graphs under SGC are also effective on other architectures.
Conclusion
• We demonstrate that removing opponent edges simultaneously can increase the
estimation errors of group influence, which might result in performance drops
• We propose a new approach to rectify graphs via multiple steps while reducing group
influence estimation errors by removing distant edges at each step
• We show that our method reduces group influence estimation errors and exhibits
superior performance compared to baselines
Thank you!

More Related Content

Similar to J. Song, ICML 2023, MLILAB, KAISTAI

VJAI Paper Reading#3-KDD2019-ClusterGCN
VJAI Paper Reading#3-KDD2019-ClusterGCNVJAI Paper Reading#3-KDD2019-ClusterGCN
VJAI Paper Reading#3-KDD2019-ClusterGCNDat Nguyen
 
220206 transformer interpretability beyond attention visualization
220206 transformer interpretability beyond attention visualization220206 transformer interpretability beyond attention visualization
220206 transformer interpretability beyond attention visualizationtaeseon ryu
 
Study and Comparison of Various Image Edge Detection Techniques
Study and Comparison of Various Image Edge Detection TechniquesStudy and Comparison of Various Image Edge Detection Techniques
Study and Comparison of Various Image Edge Detection TechniquesCSCJournals
 
Paper Presentation (Graph)
Paper Presentation (Graph)Paper Presentation (Graph)
Paper Presentation (Graph)Falguni Roy
 
An Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded SurfaceAn Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded SurfaceIJERA Editor
 
An Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded SurfaceAn Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded SurfaceIJERA Editor
 
Real time Canny edge detection
Real time Canny edge detectionReal time Canny edge detection
Real time Canny edge detectionShashank Kapoor
 
Automatic Generation of Persistent Formations Under Range Constraints
Automatic Generation of Persistent Formations Under Range ConstraintsAutomatic Generation of Persistent Formations Under Range Constraints
Automatic Generation of Persistent Formations Under Range Constraintselliando dias
 
Image segmentation methods for brain mri images
Image segmentation methods for brain mri imagesImage segmentation methods for brain mri images
Image segmentation methods for brain mri imageseSAT Journals
 
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAIJ. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAIMLILAB
 
The neural tangent link between CNN denoisers and non-local filters
The neural tangent link between CNN denoisers and non-local filtersThe neural tangent link between CNN denoisers and non-local filters
The neural tangent link between CNN denoisers and non-local filtersJulián Tachella
 
195706916 i journals-paper-template-2013
195706916 i journals-paper-template-2013195706916 i journals-paper-template-2013
195706916 i journals-paper-template-2013homeworkping3
 
Distance transforms and correlation maps for advanced 3D analysis of impact d...
Distance transforms and correlation maps for advanced 3D analysis of impact d...Distance transforms and correlation maps for advanced 3D analysis of impact d...
Distance transforms and correlation maps for advanced 3D analysis of impact d...Fabien Léonard
 
DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...
DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...
DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...Artem Lutov
 
NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...
NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...
NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...ssuser4b1f48
 
zhaowei_LIU_acme_2016
zhaowei_LIU_acme_2016zhaowei_LIU_acme_2016
zhaowei_LIU_acme_2016Zhaowei Liu
 

Similar to J. Song, ICML 2023, MLILAB, KAISTAI (20)

VJAI Paper Reading#3-KDD2019-ClusterGCN
VJAI Paper Reading#3-KDD2019-ClusterGCNVJAI Paper Reading#3-KDD2019-ClusterGCN
VJAI Paper Reading#3-KDD2019-ClusterGCN
 
220206 transformer interpretability beyond attention visualization
220206 transformer interpretability beyond attention visualization220206 transformer interpretability beyond attention visualization
220206 transformer interpretability beyond attention visualization
 
Study and Comparison of Various Image Edge Detection Techniques
Study and Comparison of Various Image Edge Detection TechniquesStudy and Comparison of Various Image Edge Detection Techniques
Study and Comparison of Various Image Edge Detection Techniques
 
NeurIPS22.pptx
NeurIPS22.pptxNeurIPS22.pptx
NeurIPS22.pptx
 
Paper Presentation (Graph)
Paper Presentation (Graph)Paper Presentation (Graph)
Paper Presentation (Graph)
 
An Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded SurfaceAn Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded Surface
 
An Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded SurfaceAn Efficient Algorithm for Edge Detection of Corroded Surface
An Efficient Algorithm for Edge Detection of Corroded Surface
 
Real time Canny edge detection
Real time Canny edge detectionReal time Canny edge detection
Real time Canny edge detection
 
Automatic Generation of Persistent Formations Under Range Constraints
Automatic Generation of Persistent Formations Under Range ConstraintsAutomatic Generation of Persistent Formations Under Range Constraints
Automatic Generation of Persistent Formations Under Range Constraints
 
Image segmentation methods for brain mri images
Image segmentation methods for brain mri imagesImage segmentation methods for brain mri images
Image segmentation methods for brain mri images
 
04 Multi-layer Feedforward Networks
04 Multi-layer Feedforward Networks04 Multi-layer Feedforward Networks
04 Multi-layer Feedforward Networks
 
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAIJ. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
 
The neural tangent link between CNN denoisers and non-local filters
The neural tangent link between CNN denoisers and non-local filtersThe neural tangent link between CNN denoisers and non-local filters
The neural tangent link between CNN denoisers and non-local filters
 
195706916 i journals-paper-template-2013
195706916 i journals-paper-template-2013195706916 i journals-paper-template-2013
195706916 i journals-paper-template-2013
 
Distance transforms and correlation maps for advanced 3D analysis of impact d...
Distance transforms and correlation maps for advanced 3D analysis of impact d...Distance transforms and correlation maps for advanced 3D analysis of impact d...
Distance transforms and correlation maps for advanced 3D analysis of impact d...
 
Image segmentation
Image segmentationImage segmentation
Image segmentation
 
DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...
DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...
DAOR - Bridging the Gap between Community and Node Representations: Graph Emb...
 
NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...
NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...
NS-CUK Joint Journal Club: S.T.Nguyen, Review on "How Attentive are Graph Att...
 
zhaowei_LIU_acme_2016
zhaowei_LIU_acme_2016zhaowei_LIU_acme_2016
zhaowei_LIU_acme_2016
 
Visibility graphs
Visibility graphsVisibility graphs
Visibility graphs
 

Recently uploaded

一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单
一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单
一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单tuuww
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单
一比一原版(UofT毕业证)多伦多大学毕业证成绩单一比一原版(UofT毕业证)多伦多大学毕业证成绩单
一比一原版(UofT毕业证)多伦多大学毕业证成绩单tuuww
 
Online resume builder management system project report.pdf
Online resume builder management system project report.pdfOnline resume builder management system project report.pdf
Online resume builder management system project report.pdfKamal Acharya
 
Top 13 Famous Civil Engineering Scientist
Top 13 Famous Civil Engineering ScientistTop 13 Famous Civil Engineering Scientist
Top 13 Famous Civil Engineering Scientistgettygaming1
 
KIT-601 Lecture Notes-UNIT-5.pdf Frame Works and Visualization
KIT-601 Lecture Notes-UNIT-5.pdf Frame Works and VisualizationKIT-601 Lecture Notes-UNIT-5.pdf Frame Works and Visualization
KIT-601 Lecture Notes-UNIT-5.pdf Frame Works and VisualizationDr. Radhey Shyam
 
Pharmacy management system project report..pdf
Pharmacy management system project report..pdfPharmacy management system project report..pdf
Pharmacy management system project report..pdfKamal Acharya
 
RESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdf
RESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdfRESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdf
RESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdfKamal Acharya
 
An improvement in the safety of big data using blockchain technology
An improvement in the safety of big data using blockchain technologyAn improvement in the safety of big data using blockchain technology
An improvement in the safety of big data using blockchain technologyBOHRInternationalJou1
 
Electrostatic field in a coaxial transmission line
Electrostatic field in a coaxial transmission lineElectrostatic field in a coaxial transmission line
Electrostatic field in a coaxial transmission lineJulioCesarSalazarHer1
 
Online book store management system project.pdf
Online book store management system project.pdfOnline book store management system project.pdf
Online book store management system project.pdfKamal Acharya
 
Paint shop management system project report.pdf
Paint shop management system project report.pdfPaint shop management system project report.pdf
Paint shop management system project report.pdfKamal Acharya
 
Teachers record management system project report..pdf
Teachers record management system project report..pdfTeachers record management system project report..pdf
Teachers record management system project report..pdfKamal Acharya
 
Dairy management system project report..pdf
Dairy management system project report..pdfDairy management system project report..pdf
Dairy management system project report..pdfKamal Acharya
 
2024 DevOps Pro Europe - Growing at the edge
2024 DevOps Pro Europe - Growing at the edge2024 DevOps Pro Europe - Growing at the edge
2024 DevOps Pro Europe - Growing at the edgePaco Orozco
 
"United Nations Park" Site Visit Report.
"United Nations Park" Site  Visit Report."United Nations Park" Site  Visit Report.
"United Nations Park" Site Visit Report.MdManikurRahman
 
Peek implant persentation - Copy (1).pdf
Peek implant persentation - Copy (1).pdfPeek implant persentation - Copy (1).pdf
Peek implant persentation - Copy (1).pdfAyahmorsy
 
RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4
RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4
RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4T.D. Shashikala
 
ONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdf
ONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdfONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdf
ONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdfKamal Acharya
 
Introduction to Machine Learning Unit-4 Notes for II-II Mechanical Engineering
Introduction to Machine Learning Unit-4 Notes for II-II Mechanical EngineeringIntroduction to Machine Learning Unit-4 Notes for II-II Mechanical Engineering
Introduction to Machine Learning Unit-4 Notes for II-II Mechanical EngineeringC Sai Kiran
 
ONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdf
ONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdfONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdf
ONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdfKamal Acharya
 

Recently uploaded (20)

一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单
一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单
一比一原版(UNK毕业证)内布拉斯加州立大学科尼分校毕业证成绩单
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单
一比一原版(UofT毕业证)多伦多大学毕业证成绩单一比一原版(UofT毕业证)多伦多大学毕业证成绩单
一比一原版(UofT毕业证)多伦多大学毕业证成绩单
 
Online resume builder management system project report.pdf
Online resume builder management system project report.pdfOnline resume builder management system project report.pdf
Online resume builder management system project report.pdf
 
Top 13 Famous Civil Engineering Scientist
Top 13 Famous Civil Engineering ScientistTop 13 Famous Civil Engineering Scientist
Top 13 Famous Civil Engineering Scientist
 
KIT-601 Lecture Notes-UNIT-5.pdf Frame Works and Visualization
KIT-601 Lecture Notes-UNIT-5.pdf Frame Works and VisualizationKIT-601 Lecture Notes-UNIT-5.pdf Frame Works and Visualization
KIT-601 Lecture Notes-UNIT-5.pdf Frame Works and Visualization
 
Pharmacy management system project report..pdf
Pharmacy management system project report..pdfPharmacy management system project report..pdf
Pharmacy management system project report..pdf
 
RESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdf
RESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdfRESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdf
RESORT MANAGEMENT AND RESERVATION SYSTEM PROJECT REPORT.pdf
 
An improvement in the safety of big data using blockchain technology
An improvement in the safety of big data using blockchain technologyAn improvement in the safety of big data using blockchain technology
An improvement in the safety of big data using blockchain technology
 
Electrostatic field in a coaxial transmission line
Electrostatic field in a coaxial transmission lineElectrostatic field in a coaxial transmission line
Electrostatic field in a coaxial transmission line
 
Online book store management system project.pdf
Online book store management system project.pdfOnline book store management system project.pdf
Online book store management system project.pdf
 
Paint shop management system project report.pdf
Paint shop management system project report.pdfPaint shop management system project report.pdf
Paint shop management system project report.pdf
 
Teachers record management system project report..pdf
Teachers record management system project report..pdfTeachers record management system project report..pdf
Teachers record management system project report..pdf
 
Dairy management system project report..pdf
Dairy management system project report..pdfDairy management system project report..pdf
Dairy management system project report..pdf
 
2024 DevOps Pro Europe - Growing at the edge
2024 DevOps Pro Europe - Growing at the edge2024 DevOps Pro Europe - Growing at the edge
2024 DevOps Pro Europe - Growing at the edge
 
"United Nations Park" Site Visit Report.
"United Nations Park" Site  Visit Report."United Nations Park" Site  Visit Report.
"United Nations Park" Site Visit Report.
 
Peek implant persentation - Copy (1).pdf
Peek implant persentation - Copy (1).pdfPeek implant persentation - Copy (1).pdf
Peek implant persentation - Copy (1).pdf
 
RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4
RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4
RM&IPR M4.pdfResearch Methodolgy & Intellectual Property Rights Series 4
 
ONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdf
ONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdfONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdf
ONLINE VEHICLE RENTAL SYSTEM PROJECT REPORT.pdf
 
Introduction to Machine Learning Unit-4 Notes for II-II Mechanical Engineering
Introduction to Machine Learning Unit-4 Notes for II-II Mechanical EngineeringIntroduction to Machine Learning Unit-4 Notes for II-II Mechanical Engineering
Introduction to Machine Learning Unit-4 Notes for II-II Mechanical Engineering
 
ONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdf
ONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdfONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdf
ONLINE CAR SERVICING SYSTEM PROJECT REPORT.pdf
 

J. Song, ICML 2023, MLILAB, KAISTAI

  • 1. RGE: A Repulsive Graph Rectification for Node Classification via Influence Jaeyun Song, Sung-Yub Kim, Eunho Yang ICML 2023 Graduate School of AI, KAIST
  • 2. Introduction ● Graph Neural Networks (GNNs) are susceptible to structural noises in graphs. ● Chen et al. (2022a) proposed an influence function for individual edges to estimate the counterfactual effects of removing them. ● We first identify the group influence estimation error of Exhaustive Group Elimination (EGE), that eliminates all opponent edges at once and propose Repulsive Group Elimination (RGE), that removes distant edges over neighboring edges to reduce group effect in GNNs. A single harmful edge (dotted) can negatively affect the predictions of multiple nodes (colored).
  • 3. Problem • Our goal is to find opponent edges 𝐸′ ⊆ 𝐸 that can improve the test performance when remove the original graph 𝐺 = (𝑉, 𝐸) . Formally, this can be formulated as • To avoid the combinatorial explosion of selecting an edge set, we consider a simple strategy, called Exhaustive edge Group Elimination (EGE): • The key assumption for the optimality of EGE is the additivity of individual influences. Graph without 𝐸′ Influence of removing 𝐸′
  • 4. Problem • Even though the influence additivity cannot be guaranteed in general graphs, we found the influence additivity can be established between distant edges for the widely used GNNs, including GCN, SGC, GAT, and GraphSAGE. • Proposition 3.1 (Influence additivity for distant edges) Let edges 𝑒, 𝑒′ ∈ 𝐸 be distant edges and, for them, assume that the removal of edge 𝑒 does not change the gradient of unaffected train node 𝒗 : for 𝑣 ∈ 𝑉𝑇𝑟 and 𝑣 ∉ 𝑉𝑇𝑟(𝑒). Then, the influence additivity holds. • Detailed proofs can be found in Appendix B!
  • 5. Method Overview • The overview of our method • Compute the influence for all edges • Remove non-adjacent edges (distant edges) with negative influence • Retrain GNNs on graphs without those edges Repeat these processes until there are no more negative edges labeled nodes unlabeled nodes non-negative edges negative edges Remove negative distant edges & Retrain GNNs Repeat processes Rectified graph
  • 6. Method • Repulsive selection rule & Multi-step edge elimination (under 1-layer SGC) labeled nodes unlabeled nodes non-negative edges negative edges Choose the most negative edge Choose the second most negative non-adjacent edge selected edges 1-st Iteration Remove edges & Retrain GNNs Choose the most negative edge 2-nd Iteration Remove edges & Retrain GNNs Repeat processes
  • 7. Experiment: Group Influence Estimation Errors • RGE significantly decreases the group influence estimation errors EGE RGE Group Influence Estimation Errors
  • 8. Experiment: Performance on Homophilous Graphs • RGE outperforms other baselines
  • 9. Experiment: Performance under Other Architectures • Rectified graphs under SGC are also effective on other architectures.
  • 10. Conclusion • We demonstrate that removing opponent edges simultaneously can increase the estimation errors of group influence, which might result in performance drops • We propose a new approach to rectify graphs via multiple steps while reducing group influence estimation errors by removing distant edges at each step • We show that our method reduces group influence estimation errors and exhibits superior performance compared to baselines