EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm
The primary objective of our research is to enhance the efficiency and effectiveness of Neural Architecture Search (NAS) with regard to Graph Neural Networks (GNNs). GNNs have emerged as powerful tools for learning from unstructured network data, compensating for several known limitations of Convolu...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/12/23/3828 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850261231319908352 |
|---|---|
| author | Younkyung Jwa Chang Wook Ahn Man-Je Kim |
| author_facet | Younkyung Jwa Chang Wook Ahn Man-Je Kim |
| author_sort | Younkyung Jwa |
| collection | DOAJ |
| description | The primary objective of our research is to enhance the efficiency and effectiveness of Neural Architecture Search (NAS) with regard to Graph Neural Networks (GNNs). GNNs have emerged as powerful tools for learning from unstructured network data, compensating for several known limitations of Convolutional Neural Networks (CNNs). However, the automatic search for optimal GNN architectures has seen little progressive advancement so far. To address this gap, we introduce the Efficient Graph Neural Architecture Search (EGNAS), a method that leverages the advantages of evolutionary search strategies. EGNAS incorporates inherited parameter sharing, allowing offspring to inherit parameters from their parents, and utilizes half epochs to improve optimization stability. In addition, EGNAS employs a combined evolutionary search, which explores both the model structure and the hyperparameters within a large search space, resulting in improved performance. Our experimental results demonstrate that EGNAS outperforms state-of-the-art methods in node classification tasks on the Cora, Citeseer, and PubMed datasets while maintaining a high degree of computational efficiency. In particular, EGNAS is the fastest GNN architecture search method in terms of search time, particularly when compared to precedently suggested evolutionary search strategies, delivering performance up to 40 times faster. |
| format | Article |
| id | doaj-art-962c190a214b4ee1b6ea593b40c3fa16 |
| institution | OA Journals |
| issn | 2227-7390 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-962c190a214b4ee1b6ea593b40c3fa162025-08-20T01:55:28ZengMDPI AGMathematics2227-73902024-12-011223382810.3390/math12233828EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary AlgorithmYounkyung Jwa0Chang Wook Ahn1Man-Je Kim2AI Graduate School, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of KoreaAI Graduate School, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of KoreaConvergence of AI, Chonnam National University, Gwangju 61186, Republic of KoreaThe primary objective of our research is to enhance the efficiency and effectiveness of Neural Architecture Search (NAS) with regard to Graph Neural Networks (GNNs). GNNs have emerged as powerful tools for learning from unstructured network data, compensating for several known limitations of Convolutional Neural Networks (CNNs). However, the automatic search for optimal GNN architectures has seen little progressive advancement so far. To address this gap, we introduce the Efficient Graph Neural Architecture Search (EGNAS), a method that leverages the advantages of evolutionary search strategies. EGNAS incorporates inherited parameter sharing, allowing offspring to inherit parameters from their parents, and utilizes half epochs to improve optimization stability. In addition, EGNAS employs a combined evolutionary search, which explores both the model structure and the hyperparameters within a large search space, resulting in improved performance. Our experimental results demonstrate that EGNAS outperforms state-of-the-art methods in node classification tasks on the Cora, Citeseer, and PubMed datasets while maintaining a high degree of computational efficiency. In particular, EGNAS is the fastest GNN architecture search method in terms of search time, particularly when compared to precedently suggested evolutionary search strategies, delivering performance up to 40 times faster.https://www.mdpi.com/2227-7390/12/23/3828graph neural networkneural architecture searchevolutionary neural architecture search |
| spellingShingle | Younkyung Jwa Chang Wook Ahn Man-Je Kim EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm Mathematics graph neural network neural architecture search evolutionary neural architecture search |
| title | EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm |
| title_full | EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm |
| title_fullStr | EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm |
| title_full_unstemmed | EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm |
| title_short | EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm |
| title_sort | egnas efficient graph neural architecture search through evolutionary algorithm |
| topic | graph neural network neural architecture search evolutionary neural architecture search |
| url | https://www.mdpi.com/2227-7390/12/23/3828 |
| work_keys_str_mv | AT younkyungjwa egnasefficientgraphneuralarchitecturesearchthroughevolutionaryalgorithm AT changwookahn egnasefficientgraphneuralarchitecturesearchthroughevolutionaryalgorithm AT manjekim egnasefficientgraphneuralarchitecturesearchthroughevolutionaryalgorithm |