Report
Control the GNN: Utilizing Neural Controller with Lyapunov Stability for Test-Time Feature Reconstruction
| العنوان: | Control the GNN: Utilizing Neural Controller with Lyapunov Stability for Test-Time Feature Reconstruction |
|---|---|
| المؤلفون: | Yang, Jielong, Ding, Rui, Ji, Feng, Wang, Hongbin, Xie, Linbo |
| سنة النشر: | 2024 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Machine Learning |
| الوصف: | The performance of graph neural networks (GNNs) is susceptible to discrepancies between training and testing sample distributions. Prior studies have attempted to enhance GNN performance by reconstructing node features during the testing phase without modifying the model parameters. However, these approaches lack theoretical analysis of the proximity between predictions and ground truth at test time. In this paper, we propose a novel node feature reconstruction method grounded in Lyapunov stability theory. Specifically, we model the GNN as a control system during the testing phase, considering node features as control variables. A neural controller that adheres to the Lyapunov stability criterion is then employed to reconstruct these node features, ensuring that the predictions progressively approach the ground truth at test time. We validate the effectiveness of our approach through extensive experiments across multiple datasets, demonstrating significant performance improvements. |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2410.09708 |
| رقم الانضمام: | edsarx.2410.09708 |
| قاعدة البيانات: | arXiv |
| ResultId |
1 |
|---|---|
| Header |
edsarx arXiv edsarx.2410.09708 1128 3 Report report 1128.01806640625 |
| PLink |
https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsarx&AN=edsarx.2410.09708&custid=s6537998&authtype=sso |
| FullText |
Array
(
[Availability] => 0
)
Array ( [0] => Array ( [Url] => http://arxiv.org/abs/2410.09708 [Name] => EDS - Arxiv [Category] => fullText [Text] => View record in Arxiv [MouseOverText] => View record in Arxiv ) ) |
| Items |
Array
(
[Name] => Title
[Label] => Title
[Group] => Ti
[Data] => Control the GNN: Utilizing Neural Controller with Lyapunov Stability for Test-Time Feature Reconstruction
)
Array ( [Name] => Author [Label] => Authors [Group] => Au [Data] => <searchLink fieldCode="AR" term="%22Yang%2C+Jielong%22">Yang, Jielong</searchLink><br /><searchLink fieldCode="AR" term="%22Ding%2C+Rui%22">Ding, Rui</searchLink><br /><searchLink fieldCode="AR" term="%22Ji%2C+Feng%22">Ji, Feng</searchLink><br /><searchLink fieldCode="AR" term="%22Wang%2C+Hongbin%22">Wang, Hongbin</searchLink><br /><searchLink fieldCode="AR" term="%22Xie%2C+Linbo%22">Xie, Linbo</searchLink> ) Array ( [Name] => DatePubCY [Label] => Publication Year [Group] => Date [Data] => 2024 ) Array ( [Name] => Subset [Label] => Collection [Group] => HoldingsInfo [Data] => Computer Science ) Array ( [Name] => Subject [Label] => Subject Terms [Group] => Su [Data] => <searchLink fieldCode="DE" term="%22Computer+Science+-+Machine+Learning%22">Computer Science - Machine Learning</searchLink> ) Array ( [Name] => Abstract [Label] => Description [Group] => Ab [Data] => The performance of graph neural networks (GNNs) is susceptible to discrepancies between training and testing sample distributions. Prior studies have attempted to enhance GNN performance by reconstructing node features during the testing phase without modifying the model parameters. However, these approaches lack theoretical analysis of the proximity between predictions and ground truth at test time. In this paper, we propose a novel node feature reconstruction method grounded in Lyapunov stability theory. Specifically, we model the GNN as a control system during the testing phase, considering node features as control variables. A neural controller that adheres to the Lyapunov stability criterion is then employed to reconstruct these node features, ensuring that the predictions progressively approach the ground truth at test time. We validate the effectiveness of our approach through extensive experiments across multiple datasets, demonstrating significant performance improvements. ) Array ( [Name] => TypeDocument [Label] => Document Type [Group] => TypDoc [Data] => Working Paper ) Array ( [Name] => URL [Label] => Access URL [Group] => URL [Data] => <link linkTarget="URL" linkTerm="http://arxiv.org/abs/2410.09708" linkWindow="_blank">http://arxiv.org/abs/2410.09708</link> ) Array ( [Name] => AN [Label] => Accession Number [Group] => ID [Data] => edsarx.2410.09708 ) |
| RecordInfo |
Array
(
[BibEntity] => Array
(
[Subjects] => Array
(
[0] => Array
(
[SubjectFull] => Computer Science - Machine Learning
[Type] => general
)
)
[Titles] => Array
(
[0] => Array
(
[TitleFull] => Control the GNN: Utilizing Neural Controller with Lyapunov Stability for Test-Time Feature Reconstruction
[Type] => main
)
)
)
[BibRelationships] => Array
(
[HasContributorRelationships] => Array
(
[0] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Yang, Jielong
)
)
)
[1] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Ding, Rui
)
)
)
[2] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Ji, Feng
)
)
)
[3] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Wang, Hongbin
)
)
)
[4] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Xie, Linbo
)
)
)
)
[IsPartOfRelationships] => Array
(
[0] => Array
(
[BibEntity] => Array
(
[Dates] => Array
(
[0] => Array
(
[D] => 12
[M] => 10
[Type] => published
[Y] => 2024
)
)
)
)
)
)
)
|
| IllustrationInfo |