Maintained by Difan Deng and Marius Lindauer.
The following list considers papers related to neural architecture search. It is by no means complete. If you miss a paper on the list, please let us know.
Please note that although NAS methods steadily improve, the quality of empirical evaluations in this field are still lagging behind compared to other areas in machine learning, AI and optimization. We would therefore like to share some best practices for empirical evaluations of NAS methods, which we believe will facilitate sustained and measurable progress in the field. If you are interested in a teaser, please read our blog post or directly jump to our checklist.
Transformers have gained increasing popularity in different domains. For a comprehensive list of papers focusing on Neural Architecture Search for Transformer-Based spaces, the awesome-transformer-search repo is all you need.
2025
3865.
Gambella, Matteo; Pittorino, Fabrizio; Roveri, Manuel
Architecture-Aware Minimization (A$^2$M): How to Find Flat Minima in Neural Architecture Search Technical Report
2025.
@techreport{gambella2025architectureawareminimizationa2mflat,
title = {Architecture-Aware Minimization (A$^2$M): How to Find Flat Minima in Neural Architecture Search},
author = {Matteo Gambella and Fabrizio Pittorino and Manuel Roveri},
url = {https://arxiv.org/abs/2503.10404},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3864.
Zhao, Tianchen; Wang, Xianpeng; Song, Xiangman
Multiobjective Backbone Network Architecture Search Based on Transfer Learning in Steel Defect Detection Journal Article
In: Neurocomputing, pp. 130012, 2025, ISSN: 0925-2312.
@article{ZHAO2025130012,
title = {Multiobjective Backbone Network Architecture Search Based on Transfer Learning in Steel Defect Detection},
author = {Tianchen Zhao and Xianpeng Wang and Xiangman Song},
url = {https://www.sciencedirect.com/science/article/pii/S0925231225006848},
doi = {https://doi.org/10.1016/j.neucom.2025.130012},
issn = {0925-2312},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Neurocomputing},
pages = {130012},
abstract = {In recent years, steel defect detection methods based on deep learning have been widely used. However, due to the shape specificity of steel defects and data scarcity, using existing convolutional neural network architectures for training requires significant expertise and time to fine-tune the hyperparameters. Transfer learning effectively tackles the challenges of data scarcity or limited computing resources by transferring domain knowledge from source tasks to related target tasks, reducing the resource consumption of model training from scratch. In this paper, we propose a transfer learning-based multiobjective backbone network architecture search method (TMBNAS). First, TMBNAS formulates defect detection network search as a multiobjective problem while optimizing its detection accuracy and model complexity. Second, an effective variable-length encoding strategy is designed to represent different building blocks and unpredictable optimal depths in convolutional neural networks, and targeted improvements are made to the crossover and mutation operators. For the specificity of the steel defect detection task, a transfer learning strategy based on similar knowledge is used to transfer the architecture and weight parameters obtained from the search in the source task to the target task, and adjust and optimize them. Finally, a dynamic adjustment mechanism based on actual constraints is designed during the search process to gradually approximate the optimal non-dominated solution set with higher detection accuracy without losing its population diversity. The proposed method is tested on the continuous casting slab and workpiece defect datasets. The experimental results show that the model searched by the proposed method can achieve better detection performance compared with manually designed deep learning algorithms and classical network architecture search methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In recent years, steel defect detection methods based on deep learning have been widely used. However, due to the shape specificity of steel defects and data scarcity, using existing convolutional neural network architectures for training requires significant expertise and time to fine-tune the hyperparameters. Transfer learning effectively tackles the challenges of data scarcity or limited computing resources by transferring domain knowledge from source tasks to related target tasks, reducing the resource consumption of model training from scratch. In this paper, we propose a transfer learning-based multiobjective backbone network architecture search method (TMBNAS). First, TMBNAS formulates defect detection network search as a multiobjective problem while optimizing its detection accuracy and model complexity. Second, an effective variable-length encoding strategy is designed to represent different building blocks and unpredictable optimal depths in convolutional neural networks, and targeted improvements are made to the crossover and mutation operators. For the specificity of the steel defect detection task, a transfer learning strategy based on similar knowledge is used to transfer the architecture and weight parameters obtained from the search in the source task to the target task, and adjust and optimize them. Finally, a dynamic adjustment mechanism based on actual constraints is designed during the search process to gradually approximate the optimal non-dominated solution set with higher detection accuracy without losing its population diversity. The proposed method is tested on the continuous casting slab and workpiece defect datasets. The experimental results show that the model searched by the proposed method can achieve better detection performance compared with manually designed deep learning algorithms and classical network architecture search methods.
3863.
Xie, Xiaofeng; Gao, Yuelin; Zhang, Yuming
An improved Artificial Protozoa Optimizer for CNN architecture optimization Journal Article
In: Neural Networks, pp. 107368, 2025, ISSN: 0893-6080.
@article{XIE2025107368,
title = {An improved Artificial Protozoa Optimizer for CNN architecture optimization},
author = {Xiaofeng Xie and Yuelin Gao and Yuming Zhang},
url = {https://www.sciencedirect.com/science/article/pii/S0893608025002473},
doi = {https://doi.org/10.1016/j.neunet.2025.107368},
issn = {0893-6080},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Neural Networks},
pages = {107368},
abstract = {In this paper, we propose a novel neural architecture search (NAS) method called MAPOCNN, which leverages an enhanced version of the Artificial Protozoa Optimizer (APO) to optimize the architecture of Convolutional Neural Networks (CNNs). The APO is known for its rapid convergence, high stability, and minimal parameter involvement. To further improve its performance, we introduce MAPO (Modified Artificial Protozoa Optimizer), which incorporates the phototaxis behavior of protozoa. This addition helps mitigate the risk of premature convergence, allowing the algorithm to explore a broader range of possible CNN architectures and ultimately identify more optimal solutions. Through rigorous experimentation on benchmark datasets, including Rectangle and Mnist-random, we demonstrate that MAPOCNN not only achieves faster convergence times but also performs competitively when compared to other state-of-the-art NAS algorithms. The results highlight the effectiveness of MAPOCNN in efficiently discovering CNN architectures that outperform existing methods in terms of both speed and accuracy. This work presents a promising direction for optimizing deep learning architectures using biologically inspired optimization techniques.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper, we propose a novel neural architecture search (NAS) method called MAPOCNN, which leverages an enhanced version of the Artificial Protozoa Optimizer (APO) to optimize the architecture of Convolutional Neural Networks (CNNs). The APO is known for its rapid convergence, high stability, and minimal parameter involvement. To further improve its performance, we introduce MAPO (Modified Artificial Protozoa Optimizer), which incorporates the phototaxis behavior of protozoa. This addition helps mitigate the risk of premature convergence, allowing the algorithm to explore a broader range of possible CNN architectures and ultimately identify more optimal solutions. Through rigorous experimentation on benchmark datasets, including Rectangle and Mnist-random, we demonstrate that MAPOCNN not only achieves faster convergence times but also performs competitively when compared to other state-of-the-art NAS algorithms. The results highlight the effectiveness of MAPOCNN in efficiently discovering CNN architectures that outperform existing methods in terms of both speed and accuracy. This work presents a promising direction for optimizing deep learning architectures using biologically inspired optimization techniques.
3862.
Onzo, Bernard-marie; Xue, Yu; Neri, Ferrante
Surrogate-assisted evolutionary neural architecture search based on smart-block discovery Journal Article
In: Expert Systems with Applications, vol. 277, pp. 127237, 2025, ISSN: 0957-4174.
@article{ONZO2025127237,
title = {Surrogate-assisted evolutionary neural architecture search based on smart-block discovery},
author = {Bernard-marie Onzo and Yu Xue and Ferrante Neri},
url = {https://www.sciencedirect.com/science/article/pii/S0957417425008590},
doi = {https://doi.org/10.1016/j.eswa.2025.127237},
issn = {0957-4174},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Expert Systems with Applications},
volume = {277},
pages = {127237},
abstract = {Neural architecture search (NAS) has emerged as a powerful method for automating neural network design, yet its high computational cost remains a significant challenge. This paper introduces hybrid training-less neural architecture search (HYTES-NAS), a novel hybrid NAS framework that integrates evolutionary computation with a training-free evaluation strategy, significantly reducing computational demands while maintaining high search efficiency. Unlike conventional NAS methods that rely on full model training, HYTES-NAS leverages a surrogate-assisted scoring mechanism to assess candidate architectures efficiently. Additionally, a smart-block discovery strategy and particle swarm optimisation are employed to refine the search space and accelerate convergence. Experimental results on multiple NAS benchmarks demonstrate that HYTES-NAS achieves superior performance with significantly lower computational cost compared to state-of-the-art NAS methods. This work provides a promising and scalable solution for efficient NAS, making high-performance architecture search more accessible for real-world applications.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Neural architecture search (NAS) has emerged as a powerful method for automating neural network design, yet its high computational cost remains a significant challenge. This paper introduces hybrid training-less neural architecture search (HYTES-NAS), a novel hybrid NAS framework that integrates evolutionary computation with a training-free evaluation strategy, significantly reducing computational demands while maintaining high search efficiency. Unlike conventional NAS methods that rely on full model training, HYTES-NAS leverages a surrogate-assisted scoring mechanism to assess candidate architectures efficiently. Additionally, a smart-block discovery strategy and particle swarm optimisation are employed to refine the search space and accelerate convergence. Experimental results on multiple NAS benchmarks demonstrate that HYTES-NAS achieves superior performance with significantly lower computational cost compared to state-of-the-art NAS methods. This work provides a promising and scalable solution for efficient NAS, making high-performance architecture search more accessible for real-world applications.
3861.
Jeon, Jeimin; Oh, Youngmin; Lee, Junghyup; Baek, Donghyeon; Kim, Dohyung; Eom, Chanho; Ham, Bumsub
Subnet-Aware Dynamic Supernet Training for Neural Architecture Search Technical Report
2025.
@techreport{jeon2025subnetawaredynamicsupernettraining,
title = {Subnet-Aware Dynamic Supernet Training for Neural Architecture Search},
author = {Jeimin Jeon and Youngmin Oh and Junghyup Lee and Donghyeon Baek and Dohyung Kim and Chanho Eom and Bumsub Ham},
url = {https://arxiv.org/abs/2503.10740},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3860.
Winter, Benjamin David; Teahan, William J.
Ecological Neural Architecture Search Technical Report
2025.
@techreport{winter2025ecologicalneuralarchitecturesearch,
title = {Ecological Neural Architecture Search},
author = {Benjamin David Winter and William J. Teahan},
url = {https://arxiv.org/abs/2503.10908},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3859.
Winter, Benjamin David; Teahan, William John
Evaluating a Novel Neuroevolution and Neural Architecture Search System Technical Report
2025.
@techreport{winter2025evaluatingnovelneuroevolutionneural,
title = {Evaluating a Novel Neuroevolution and Neural Architecture Search System},
author = {Benjamin David Winter and William John Teahan},
url = {https://arxiv.org/abs/2503.10869},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3858.
Wang, Mansheng; Gu, Yu; Yang, Lidong; Zhang, Baohua; Wang, Jing; Lu, Xiaoqi; Li, Jianjun; Liu, Xin; Zhao, Ying; Yu, Dahua; Tang, Siyuan; He, Qun
A novel high-precision bilevel optimization method for 3D pulmonary nodule classification Journal Article
In: Physica Medica, vol. 133, pp. 104954, 2025, ISSN: 1120-1797.
@article{WANG2025104954,
title = {A novel high-precision bilevel optimization method for 3D pulmonary nodule classification},
author = {Mansheng Wang and Yu Gu and Lidong Yang and Baohua Zhang and Jing Wang and Xiaoqi Lu and Jianjun Li and Xin Liu and Ying Zhao and Dahua Yu and Siyuan Tang and Qun He},
url = {https://www.sciencedirect.com/science/article/pii/S112017972500064X},
doi = {https://doi.org/10.1016/j.ejmp.2025.104954},
issn = {1120-1797},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Physica Medica},
volume = {133},
pages = {104954},
abstract = {Background and objective
Classification of pulmonary nodules is important for the early diagnosis of lung cancer; however, the manual design of classification models requires substantial expert effort. To automate the model design process, we propose a neural architecture search with high-precision bilevel optimization (NAS-HBO) that directly searches for the optimal network on three-dimensional (3D) images.
Methods
We propose a novel high-precision bilevel optimization method (HBOM) to search for an optimal 3D pulmonary nodule classification model. We employed memory optimization techniques with a partially decoupled operation-weighting method to reduce the memory overhead while maintaining path selection stability. Additionally, we introduce a novel maintaining receptive field criterion (MRFC) within the NAS-HBO framework. MRFC narrows the search space by selecting and expanding the 3D Mobile Inverted Residual Bottleneck Block (3D-MBconv) operation based on previous receptive fields, thereby enhancing the scalability and practical application capabilities of NAS-HBO in terms of model complexity and performance.
Results
In this study, 888 CT images, including 554 benign and 450 malignant nodules, were obtained from the LIDC-IDRI dataset. The results showed that NAS-HBO achieved an impressive accuracy of 91.51 % after less than 6 h of searching, utilizing a mere 12.79 M parameters.
Conclusion
The proposed NAS-HBO method effectively automates the design of 3D pulmonary nodule classification models, achieving impressive accuracy with efficient parameters. By incorporating the HBOM and MRFC techniques, we demonstrated enhanced accuracy and scalability in model optimization for early lung cancer diagnosis. The related codes and results have been released at https://github.com/GuYuIMUST/NAS-HBO.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Background and objective
Classification of pulmonary nodules is important for the early diagnosis of lung cancer; however, the manual design of classification models requires substantial expert effort. To automate the model design process, we propose a neural architecture search with high-precision bilevel optimization (NAS-HBO) that directly searches for the optimal network on three-dimensional (3D) images.
Methods
We propose a novel high-precision bilevel optimization method (HBOM) to search for an optimal 3D pulmonary nodule classification model. We employed memory optimization techniques with a partially decoupled operation-weighting method to reduce the memory overhead while maintaining path selection stability. Additionally, we introduce a novel maintaining receptive field criterion (MRFC) within the NAS-HBO framework. MRFC narrows the search space by selecting and expanding the 3D Mobile Inverted Residual Bottleneck Block (3D-MBconv) operation based on previous receptive fields, thereby enhancing the scalability and practical application capabilities of NAS-HBO in terms of model complexity and performance.
Results
In this study, 888 CT images, including 554 benign and 450 malignant nodules, were obtained from the LIDC-IDRI dataset. The results showed that NAS-HBO achieved an impressive accuracy of 91.51 % after less than 6 h of searching, utilizing a mere 12.79 M parameters.
Conclusion
The proposed NAS-HBO method effectively automates the design of 3D pulmonary nodule classification models, achieving impressive accuracy with efficient parameters. By incorporating the HBOM and MRFC techniques, we demonstrated enhanced accuracy and scalability in model optimization for early lung cancer diagnosis. The related codes and results have been released at https://github.com/GuYuIMUST/NAS-HBO.
Classification of pulmonary nodules is important for the early diagnosis of lung cancer; however, the manual design of classification models requires substantial expert effort. To automate the model design process, we propose a neural architecture search with high-precision bilevel optimization (NAS-HBO) that directly searches for the optimal network on three-dimensional (3D) images.
Methods
We propose a novel high-precision bilevel optimization method (HBOM) to search for an optimal 3D pulmonary nodule classification model. We employed memory optimization techniques with a partially decoupled operation-weighting method to reduce the memory overhead while maintaining path selection stability. Additionally, we introduce a novel maintaining receptive field criterion (MRFC) within the NAS-HBO framework. MRFC narrows the search space by selecting and expanding the 3D Mobile Inverted Residual Bottleneck Block (3D-MBconv) operation based on previous receptive fields, thereby enhancing the scalability and practical application capabilities of NAS-HBO in terms of model complexity and performance.
Results
In this study, 888 CT images, including 554 benign and 450 malignant nodules, were obtained from the LIDC-IDRI dataset. The results showed that NAS-HBO achieved an impressive accuracy of 91.51 % after less than 6 h of searching, utilizing a mere 12.79 M parameters.
Conclusion
The proposed NAS-HBO method effectively automates the design of 3D pulmonary nodule classification models, achieving impressive accuracy with efficient parameters. By incorporating the HBOM and MRFC techniques, we demonstrated enhanced accuracy and scalability in model optimization for early lung cancer diagnosis. The related codes and results have been released at https://github.com/GuYuIMUST/NAS-HBO.
3857.
Xue, Yu; Hu, Bohan; Neri, Ferrante
A Surrogate Model With Multiple Comparisons and Semi-Online Learning for Evolutionary Neural Architecture Search Journal Article
In: IEEE Transactions on Emerging Topics in Computational Intelligence, pp. 1-13, 2025.
@article{10935345,
title = {A Surrogate Model With Multiple Comparisons and Semi-Online Learning for Evolutionary Neural Architecture Search},
author = {Yu Xue and Bohan Hu and Ferrante Neri},
url = {https://ieeexplore.ieee.org/abstract/document/10935345},
doi = {10.1109/TETCI.2025.3547621},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Emerging Topics in Computational Intelligence},
pages = {1-13},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3856.
Li, Kefan; Wan, Yuting; Ma, Ailong; Zhong, Yanfei
A Lightweight Multi-Scale and Multi-Attention Hyperspectral Image Classification Network Based on Multi-Stage Search Journal Article
In: IEEE Transactions on Geoscience and Remote Sensing, pp. 1-1, 2025.
@article{10935661,
title = {A Lightweight Multi-Scale and Multi-Attention Hyperspectral Image Classification Network Based on Multi-Stage Search},
author = {Kefan Li and Yuting Wan and Ailong Ma and Yanfei Zhong},
url = {https://ieeexplore.ieee.org/abstract/document/10935661},
doi = {10.1109/TGRS.2025.3553147},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3855.
Xue, Xin; Zhou, Haoyi; Chen, Tianyu; Zhang, Shuai; Long, Yizhou; Li, Jianxin
Instructing the Architecture Search for Spatial-temporal Sequence Forecasting with LLM Technical Report
2025.
@techreport{xue2025instructingarchitecturesearchspatialtemporal,
title = {Instructing the Architecture Search for Spatial-temporal Sequence Forecasting with LLM},
author = {Xin Xue and Haoyi Zhou and Tianyu Chen and Shuai Zhang and Yizhou Long and Jianxin Li},
url = {https://arxiv.org/abs/2503.17994},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3854.
Chen, Zhen-Song; Ding, Hong-Wei; Wang, Xian-Jia; Pedrycz, Witold
ZeroLM: Data-Free Transformer Architecture Search for Language Models Technical Report
2025.
@techreport{chen2025zerolmdatafreetransformerarchitecture,
title = {ZeroLM: Data-Free Transformer Architecture Search for Language Models},
author = {Zhen-Song Chen and Hong-Wei Ding and Xian-Jia Wang and Witold Pedrycz},
url = {https://arxiv.org/abs/2503.18646},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3853.
Alotaibi, Abrar; Ahmed, Moataz
Neural Architecture Search for Generative Adversarial Networks: A Comprehensive Review and Critical Analysis Journal Article
In: Applied Sciences, vol. 15, no. 7, 2025, ISSN: 2076-3417.
@article{app15073623,
title = {Neural Architecture Search for Generative Adversarial Networks: A Comprehensive Review and Critical Analysis},
author = {Abrar Alotaibi and Moataz Ahmed},
url = {https://www.mdpi.com/2076-3417/15/7/3623},
doi = {10.3390/app15073623},
issn = {2076-3417},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Applied Sciences},
volume = {15},
number = {7},
abstract = {Neural Architecture Search (NAS) has emerged as a pivotal technique in optimizing the design of Generative Adversarial Networks (GANs), automating the search for effective architectures while addressing the challenges inherent in manual design. This paper provides a comprehensive review of NAS methods applied to GANs, categorizing and comparing various approaches based on criteria such as search strategies, evaluation metrics, and performance outcomes. The review highlights the benefits of NAS in improving GAN performance, stability, and efficiency, while also identifying limitations and areas for future research. Key findings include the superiority of evolutionary algorithms and gradient-based methods in certain contexts, the importance of robust evaluation metrics beyond traditional scores like Inception Score (IS) and Fréchet Inception Distance (FID), and the need for diverse datasets in assessing GAN performance. By presenting a structured comparison of existing NAS-GAN techniques, this paper aims to guide researchers in developing more effective NAS methods and advancing the field of GANs.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Neural Architecture Search (NAS) has emerged as a pivotal technique in optimizing the design of Generative Adversarial Networks (GANs), automating the search for effective architectures while addressing the challenges inherent in manual design. This paper provides a comprehensive review of NAS methods applied to GANs, categorizing and comparing various approaches based on criteria such as search strategies, evaluation metrics, and performance outcomes. The review highlights the benefits of NAS in improving GAN performance, stability, and efficiency, while also identifying limitations and areas for future research. Key findings include the superiority of evolutionary algorithms and gradient-based methods in certain contexts, the importance of robust evaluation metrics beyond traditional scores like Inception Score (IS) and Fréchet Inception Distance (FID), and the need for diverse datasets in assessing GAN performance. By presenting a structured comparison of existing NAS-GAN techniques, this paper aims to guide researchers in developing more effective NAS methods and advancing the field of GANs.
3852.
Zhu, Lixian; Wang, Su; Jin, Xiaokun; Zheng, Kai; Zhang, Jian; Sun, Shuting; Tian, Fuze; Cai, Ran; Hu, Bin
MDH-NAS: Accelerating EEG Signal Classification with Mixed-Level Differentiable and Hardware-Aware Neural Architecture Search Journal Article
In: IEEE Internet of Things Journal, pp. 1-1, 2025.
@article{10938080,
title = {MDH-NAS: Accelerating EEG Signal Classification with Mixed-Level Differentiable and Hardware-Aware Neural Architecture Search},
author = {Lixian Zhu and Su Wang and Xiaokun Jin and Kai Zheng and Jian Zhang and Shuting Sun and Fuze Tian and Ran Cai and Bin Hu},
url = {https://ieeexplore.ieee.org/abstract/document/10938080},
doi = {10.1109/JIOT.2025.3553450},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Internet of Things Journal},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3851.
Tong, Lyuyang; Liu, Jie; Du, Bo
SceneFormer: Neural Architecture Search of Transformers for Remote Sensing Scene Classification Journal Article
In: IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1-15, 2025.
@article{10942436,
title = {SceneFormer: Neural Architecture Search of Transformers for Remote Sensing Scene Classification},
author = {Lyuyang Tong and Jie Liu and Bo Du},
url = {https://ieeexplore.ieee.org/abstract/document/10942436},
doi = {10.1109/TGRS.2025.3555207},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
volume = {63},
pages = {1-15},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3850.
Chen, Oscal Tzyh-Chiang; Chang, Yu-Xuan; Chung, Chih-Yu; Cheng, Ya-Yun; HA, Manh-Hung
Hardware-Aware Iterative One-Shot Neural Architecture Search With Adaptable Knowledge Distillation for Efficient Edge Computing Journal Article
In: IEEE Access, vol. 13, pp. 54204-54222, 2025.
@article{10938148,
title = {Hardware-Aware Iterative One-Shot Neural Architecture Search With Adaptable Knowledge Distillation for Efficient Edge Computing},
author = {Oscal Tzyh-Chiang Chen and Yu-Xuan Chang and Chih-Yu Chung and Ya-Yun Cheng and Manh-Hung HA},
url = {https://ieeexplore.ieee.org/document/10938148},
doi = {10.1109/ACCESS.2025.3554185},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Access},
volume = {13},
pages = {54204-54222},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3849.
Huang, Junhao; Xue, Bing; Sun, Yanan; Zhang, Mengjie; Yen, Gary G.
Automated design of neural networks with multi-scale convolutions via multi-path weight sampling Journal Article
In: Pattern Recognition, vol. 165, pp. 111605, 2025, ISSN: 0031-3203.
@article{HUANG2025111605,
title = {Automated design of neural networks with multi-scale convolutions via multi-path weight sampling},
author = {Junhao Huang and Bing Xue and Yanan Sun and Mengjie Zhang and Gary G. Yen},
url = {https://www.sciencedirect.com/science/article/pii/S0031320325002651},
doi = {https://doi.org/10.1016/j.patcog.2025.111605},
issn = {0031-3203},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Pattern Recognition},
volume = {165},
pages = {111605},
abstract = {The performance of convolutional neural networks (CNNs) relies heavily on the architecture design. Recently, an increasingly prevalent trend in CNN architecture design is the utilization of ingeniously crafted building blocks, e.g., the MixConv module, for improving the model expressivity and efficiency. To leverage the feature learning capability of multi-scale convolution while further reducing its computational complexity, this paper presents a computationally efficient yet powerful module, dubbed EMixConv, by combining parameter-free concatenation-based feature reuse with multi-scale convolution. In addition, we propose a one-shot neural architecture search (NAS) method integrating the EMixConv module to automatically search for the optimal combination of the related architectural parameters. Furthermore, an efficient multi-path weight sampling mechanism is developed to enhance the robustness of weight inheritance in the supernet. We demonstrate the effectiveness of the proposed module and the NAS algorithm on three popular image classification tasks. The developed models, dubbed EMixNets, outperform most state-of-the-art architectures with fewer parameters and computations on the CIFAR datasets. On ImageNet, EMixNet is superior to a majority of compared methods and is also more compact and computationally efficient.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The performance of convolutional neural networks (CNNs) relies heavily on the architecture design. Recently, an increasingly prevalent trend in CNN architecture design is the utilization of ingeniously crafted building blocks, e.g., the MixConv module, for improving the model expressivity and efficiency. To leverage the feature learning capability of multi-scale convolution while further reducing its computational complexity, this paper presents a computationally efficient yet powerful module, dubbed EMixConv, by combining parameter-free concatenation-based feature reuse with multi-scale convolution. In addition, we propose a one-shot neural architecture search (NAS) method integrating the EMixConv module to automatically search for the optimal combination of the related architectural parameters. Furthermore, an efficient multi-path weight sampling mechanism is developed to enhance the robustness of weight inheritance in the supernet. We demonstrate the effectiveness of the proposed module and the NAS algorithm on three popular image classification tasks. The developed models, dubbed EMixNets, outperform most state-of-the-art architectures with fewer parameters and computations on the CIFAR datasets. On ImageNet, EMixNet is superior to a majority of compared methods and is also more compact and computationally efficient.
3848.
Gupta, Vyom Kumar; Yadav, Abhishek; Kumar, Mirgender; Kumar, Binod; Sunny,
On-Chip Implementation of Neural Network-Based Classifier Models for E-Nose With Chemometric Analysis Journal Article
In: IEEE Transactions on Instrumentation and Measurement, vol. 74, pp. 1-8, 2025.
@article{10938683,
title = {On-Chip Implementation of Neural Network-Based Classifier Models for E-Nose With Chemometric Analysis},
author = {Vyom Kumar Gupta and Abhishek Yadav and Mirgender Kumar and Binod Kumar and Sunny},
url = {https://ieeexplore.ieee.org/abstract/document/10938683},
doi = {10.1109/TIM.2025.3554292},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Instrumentation and Measurement},
volume = {74},
pages = {1-8},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3847.
Lin, Yuxin; Zhu, Chaoyang
Detection of Pedestrian Movement Poses in High-Speed Autonomous Driving Environments Using DVS Proceedings Article
In: Yang, Jie; Zheng, Yuanjie; Gong, Chen (Ed.): Pattern Analysis and Machine Intelligence, pp. 54–66, Springer Nature Singapore, Singapore, 2025, ISBN: 978-981-96-3349-4.
@inproceedings{10.1007/978-981-96-3349-4_8,
title = {Detection of Pedestrian Movement Poses in High-Speed Autonomous Driving Environments Using DVS},
author = {Yuxin Lin and Chaoyang Zhu},
editor = {Jie Yang and Yuanjie Zheng and Chen Gong},
url = {https://link.springer.com/chapter/10.1007/978-981-96-3349-4_8},
isbn = {978-981-96-3349-4},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {Pattern Analysis and Machine Intelligence},
pages = {54–66},
publisher = {Springer Nature Singapore},
address = {Singapore},
abstract = {In the realm of autonomous driving, the detection and prediction of pedestrian movement poses at high speeds are crucial for enhancing vehicular safety. Traditional imaging technologies, while rich in detail, suffer from limitations such as low frame rates and shutter-induced latencies, which can impede the rapid detection necessary in high-speed environments. This paper introduces a novel algorithm that leverages the capabilities of Dynamic Vision Sensors (DVS) to detect pedestrian poses under high-speed conditions. Unlike conventional cameras, DVS operate on the principle of capturing changes in light intensity at each pixel, allowing for data generation with high temporal resolution and minimal latency. Our approach integrates this technology with a Neural Architecture Search (NAS) optimized version of the YOLO-NAS model, specifically adapted to process the unique event-based data produced by DVS. This integration not only enhances the detection capabilities but also significantly reduces the system's response time. Comparative experiments demonstrate that our DVS-based system achieves a mean Average Precision (mAP) of 85.4%. These results underscore the potential of event-based vision sensors in transforming pedestrian detection frameworks within autonomous driving systems, offering substantial improvements in both accuracy and speed.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In the realm of autonomous driving, the detection and prediction of pedestrian movement poses at high speeds are crucial for enhancing vehicular safety. Traditional imaging technologies, while rich in detail, suffer from limitations such as low frame rates and shutter-induced latencies, which can impede the rapid detection necessary in high-speed environments. This paper introduces a novel algorithm that leverages the capabilities of Dynamic Vision Sensors (DVS) to detect pedestrian poses under high-speed conditions. Unlike conventional cameras, DVS operate on the principle of capturing changes in light intensity at each pixel, allowing for data generation with high temporal resolution and minimal latency. Our approach integrates this technology with a Neural Architecture Search (NAS) optimized version of the YOLO-NAS model, specifically adapted to process the unique event-based data produced by DVS. This integration not only enhances the detection capabilities but also significantly reduces the system's response time. Comparative experiments demonstrate that our DVS-based system achieves a mean Average Precision (mAP) of 85.4%. These results underscore the potential of event-based vision sensors in transforming pedestrian detection frameworks within autonomous driving systems, offering substantial improvements in both accuracy and speed.
3846.
Hoang, Anh Tuan; Viharos, Zsolt János
Model Input-Output Configuration Search With Embedded Feature Selection for Sensor Time-Series and Image Classification Journal Article
In: IEEE Access, vol. 13, pp. 58960-58977, 2025.
@article{10943129,
title = {Model Input-Output Configuration Search With Embedded Feature Selection for Sensor Time-Series and Image Classification},
author = {Anh Tuan Hoang and Zsolt János Viharos},
url = {Model Input-Output Configuration Search With Embedded Feature Selection for Sensor Time-Series and Image Classification},
doi = {10.1109/ACCESS.2025.3555379},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Access},
volume = {13},
pages = {58960-58977},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3845.
Mendis, Hashan Roshantha; Yen, Chih-Hsuan; Kang, Chih-Kai; Hsiu, Pi-Cheng
Intermittent-Friendly Neural Architecture Search: Demystifying Accuracy and Overhead Trade-Offs Journal Article
In: IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, pp. 1-1, 2025.
@article{10944793,
title = {Intermittent-Friendly Neural Architecture Search: Demystifying Accuracy and Overhead Trade-Offs},
author = {Hashan Roshantha Mendis and Chih-Hsuan Yen and Chih-Kai Kang and Pi-Cheng Hsiu},
url = {https://ieeexplore.ieee.org/abstract/document/10944793},
doi = {10.1109/TCAD.2025.3555963},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3844.
Poddenige, Deshani Geethika; Seneviratne, Sachith; Senanayake, Damith; Niranjan, Mahesan; Suganthan, PN; Halgamuge, Saman
Arch-LLM: Taming LLMs for Neural Architecture Generation via Unsupervised Discrete Representation Learning Technical Report
2025.
@techreport{poddenige2025archllmtamingllmsneural,
title = {Arch-LLM: Taming LLMs for Neural Architecture Generation via Unsupervised Discrete Representation Learning},
author = {Deshani Geethika Poddenige and Sachith Seneviratne and Damith Senanayake and Mahesan Niranjan and PN Suganthan and Saman Halgamuge},
url = {https://arxiv.org/abs/2503.22063},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3843.
Abdennadher, Yesmine; Perin, Giovanni; Mazzieri, Riccardo; Pegoraro, Jacopo; Rossi, Michele
LightSNN: Lightweight Architecture Search for Sparse and Accurate Spiking Neural Networks Technical Report
2025.
@techreport{abdennadher2025lightsnnlightweightarchitecturesearch,
title = {LightSNN: Lightweight Architecture Search for Sparse and Accurate Spiking Neural Networks},
author = {Yesmine Abdennadher and Giovanni Perin and Riccardo Mazzieri and Jacopo Pegoraro and Michele Rossi},
url = {https://arxiv.org/abs/2503.21846},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3842.
Tran, Tony; Hu, Bin
FACETS: Efficient Once-for-all Object Detection via Constrained Iterative Search Technical Report
2025.
@techreport{tran2025facetsefficientonceforallobject,
title = {FACETS: Efficient Once-for-all Object Detection via Constrained Iterative Search},
author = {Tony Tran and Bin Hu},
url = {https://arxiv.org/abs/2503.21999},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3841.
Mbamba, Christian Kazadi; Keymer, Philip; Alvi, Maira; Topalian, Sebastian O. N.; Din, Fareed Ud; Batstone, Damien J.
Enhancing data quality in wastewater processes: Missing data imputation with deep Variational Autoencoders and genetic algorithms Journal Article
In: Computers & Chemical Engineering, vol. 199, pp. 109123, 2025, ISSN: 0098-1354.
@article{KAZADIMBAMBA2025109123,
title = {Enhancing data quality in wastewater processes: Missing data imputation with deep Variational Autoencoders and genetic algorithms},
author = {Christian Kazadi Mbamba and Philip Keymer and Maira Alvi and Sebastian O. N. Topalian and Fareed Ud Din and Damien J. Batstone},
url = {https://www.sciencedirect.com/science/article/pii/S0098135425001279},
doi = {https://doi.org/10.1016/j.compchemeng.2025.109123},
issn = {0098-1354},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Computers & Chemical Engineering},
volume = {199},
pages = {109123},
abstract = {Missing data is a persistent challenge in wastewater analysis, often leading to biased results and reduced accuracy. This study introduces an innovative Automated Machine Learning (AutoML) framework that combines deep learning-based variational autoencoders (VAEs) and genetic algorithms (GAs) to address this issue. VAEs are employed to impute missing values by learning latent data representations, while GAs optimize the VAE architecture and hyperparameters, including the size of the latent space. The framework is specifically designed to handle the complex and nonlinear relationships in wastewater datasets. The framework was trained and validated using data from a full-scale water resource recovery facility. The imputed data from the optimized VAE, developed using the GA-based AutoML framework, is then used to train predictive models. Experimental evaluations demonstrate the effectiveness of the proposed approach over traditional imputation methods. The results reveal that the models can accurately predict key variables such as ammonia nitrogen (NH4-N), nitrate nitrogen (NO3-N), pH, and biogas flow rate, using imputed data. The scalability and adaptability of this framework make it valuable for real-time wastewater monitoring and predictive analytics.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Missing data is a persistent challenge in wastewater analysis, often leading to biased results and reduced accuracy. This study introduces an innovative Automated Machine Learning (AutoML) framework that combines deep learning-based variational autoencoders (VAEs) and genetic algorithms (GAs) to address this issue. VAEs are employed to impute missing values by learning latent data representations, while GAs optimize the VAE architecture and hyperparameters, including the size of the latent space. The framework is specifically designed to handle the complex and nonlinear relationships in wastewater datasets. The framework was trained and validated using data from a full-scale water resource recovery facility. The imputed data from the optimized VAE, developed using the GA-based AutoML framework, is then used to train predictive models. Experimental evaluations demonstrate the effectiveness of the proposed approach over traditional imputation methods. The results reveal that the models can accurately predict key variables such as ammonia nitrogen (NH4-N), nitrate nitrogen (NO3-N), pH, and biogas flow rate, using imputed data. The scalability and adaptability of this framework make it valuable for real-time wastewater monitoring and predictive analytics.
3840.
Zhu, Qingling; Yang, Yeming; Liu, Songbai; Lin, Qiuzhen; Tan, Kay Chen
SCGAN: Sampling and Clustering-Based Neural Architecture Search for GANs Journal Article
In: IEEE Transactions on Emerging Topics in Computational Intelligence, pp. 1-12, 2025.
@article{10944782,
title = {SCGAN: Sampling and Clustering-Based Neural Architecture Search for GANs},
author = {Qingling Zhu and Yeming Yang and Songbai Liu and Qiuzhen Lin and Kay Chen Tan},
url = {https://ieeexplore.ieee.org/abstract/document/10944782},
doi = {10.1109/TETCI.2025.3547611},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Emerging Topics in Computational Intelligence},
pages = {1-12},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3839.
Zhang, TingJie; Liu, HaiLin
Multi-Task Neural Architecture Search Using Architecture Embedding and Transfer Rank Technical Report
2025.
@techreport{zhang2025multitaskneuralarchitecturesearch,
title = {Multi-Task Neural Architecture Search Using Architecture Embedding and Transfer Rank},
author = {TingJie Zhang and HaiLin Liu},
url = {https://arxiv.org/abs/2504.00772},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3838.
Yamasaki, Tomomasa; Wang, Zhehui; Luo, Tao; Chen, Niangjun; Wang, Bo
RBFleX-NAS: Training-Free Neural Architecture Search Using Radial Basis Function Kernel and Hyperparameter Detection Technical Report
2025.
@techreport{yamasaki2025rbflexnastrainingfreeneuralarchitecture,
title = {RBFleX-NAS: Training-Free Neural Architecture Search Using Radial Basis Function Kernel and Hyperparameter Detection},
author = {Tomomasa Yamasaki and Zhehui Wang and Tao Luo and Niangjun Chen and Bo Wang},
url = {https://arxiv.org/abs/2503.22733},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3837.
Peri, Gianluca; Giambagli, Lorenzo; Chicchi, Lorenzo; Fanelli, Duccio
Spectral Architecture Search for Neural Networks Technical Report
2025.
@techreport{peri2025spectralarchitecturesearchneural,
title = {Spectral Architecture Search for Neural Networks},
author = {Gianluca Peri and Lorenzo Giambagli and Lorenzo Chicchi and Duccio Fanelli},
url = {https://arxiv.org/abs/2504.00885},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3836.
Capuano, Francesco; Tiboni, Gabriele; Cavagnero, Niccolò; Averta, Giuseppe
Sim-is-More: Randomizing HW-NAS with Synthetic Devices Technical Report
2025.
@techreport{capuano2025simismorerandomizinghwnassynthetic,
title = {Sim-is-More: Randomizing HW-NAS with Synthetic Devices},
author = {Francesco Capuano and Gabriele Tiboni and Niccolò Cavagnero and Giuseppe Averta},
url = {https://arxiv.org/abs/2504.00663},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3835.
Yu, YiMing; Zutty, Jason
LLM-Guided Evolution: An Autonomous Model Optimization for Object Detection Technical Report
2025.
@techreport{yu2025llmguidedevolutionautonomousmodel,
title = {LLM-Guided Evolution: An Autonomous Model Optimization for Object Detection},
author = {YiMing Yu and Jason Zutty},
url = {https://arxiv.org/abs/2504.02280},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3834.
Roshtkhari, Mehraveh Javan; Toews, Matthew; Pedersoli, Marco
Neural Architecture Search by Learning a Hierarchical Search Space Technical Report
2025.
@techreport{roshtkhari2025neuralarchitecturesearchlearning,
title = {Neural Architecture Search by Learning a Hierarchical Search Space},
author = {Mehraveh Javan Roshtkhari and Matthew Toews and Marco Pedersoli},
url = {https://arxiv.org/abs/2503.21061},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3833.
Kang, Siyuan; Sun, Yinghao; Li, Shuguang; Xu, Yaozong; Li, Yuke; Chen, Guangjie; Xue, Fei
A lightweight neural network search algorithm based on in-place distillation and performance prediction for hardware-aware optimization Journal Article
In: Engineering Applications of Artificial Intelligence, vol. 151, pp. 110775, 2025, ISSN: 0952-1976.
@article{KANG2025110775,
title = {A lightweight neural network search algorithm based on in-place distillation and performance prediction for hardware-aware optimization},
author = {Siyuan Kang and Yinghao Sun and Shuguang Li and Yaozong Xu and Yuke Li and Guangjie Chen and Fei Xue},
url = {https://www.sciencedirect.com/science/article/pii/S0952197625007754},
doi = {https://doi.org/10.1016/j.engappai.2025.110775},
issn = {0952-1976},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Engineering Applications of Artificial Intelligence},
volume = {151},
pages = {110775},
abstract = {Due to the limited computing resources of edge devices, traditional object detection algorithms struggle to meet the efficiency and accuracy requirements of autonomous driving. Consequently, designing a neural network model that balances hardware resource requirements, operating speed, and accuracy is crucial. To address this, by integrating algorithm with hardware characteristics, we propose a lightweight neural network architecture search algorithm based on in-place distillation and performance predictor (LNIP). Initially, we focus on optimizing the operators of the you only look once version 8 nano (YOLOv8n) and dynamically adjust its network structure. Then, we trained a super-network using a progressive shrinking strategy, the sandwich rule, and in-place distillation. Subsequently, we employed a Gaussian process to model the relationship between network architecture and accuracy, utilizing encoding methods and custom kernel function to develop high-performance predictor. Finally, during the search process, we introduce a reward function based on Pareto optimality to balance the performance of the model with hardware constraints. Building upon this foundation, we design an efficient search algorithm based on the performance predictor to progressively explore the optimal network structure tailored to hardware characteristics. We compared our lightweight network with state-of-the-art methods on the BDD100K, COCO, and PASCAL VOC datasets and deployed it on the Black Sesame A1000 and NVIDIA Xavier for comprehensive evaluation. On the NVIDIA Xavier, the lightweight network achieves a latency of 11.81 ms and an edge precision of 46.1 %. These experimental results demonstrate that our method outperforms existing methods in balancing hardware constraints and model performance.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Due to the limited computing resources of edge devices, traditional object detection algorithms struggle to meet the efficiency and accuracy requirements of autonomous driving. Consequently, designing a neural network model that balances hardware resource requirements, operating speed, and accuracy is crucial. To address this, by integrating algorithm with hardware characteristics, we propose a lightweight neural network architecture search algorithm based on in-place distillation and performance predictor (LNIP). Initially, we focus on optimizing the operators of the you only look once version 8 nano (YOLOv8n) and dynamically adjust its network structure. Then, we trained a super-network using a progressive shrinking strategy, the sandwich rule, and in-place distillation. Subsequently, we employed a Gaussian process to model the relationship between network architecture and accuracy, utilizing encoding methods and custom kernel function to develop high-performance predictor. Finally, during the search process, we introduce a reward function based on Pareto optimality to balance the performance of the model with hardware constraints. Building upon this foundation, we design an efficient search algorithm based on the performance predictor to progressively explore the optimal network structure tailored to hardware characteristics. We compared our lightweight network with state-of-the-art methods on the BDD100K, COCO, and PASCAL VOC datasets and deployed it on the Black Sesame A1000 and NVIDIA Xavier for comprehensive evaluation. On the NVIDIA Xavier, the lightweight network achieves a latency of 11.81 ms and an edge precision of 46.1 %. These experimental results demonstrate that our method outperforms existing methods in balancing hardware constraints and model performance.
3832.
Chehade, Adel; Ragusa, Edoardo; Gastaldo, Paolo; Zunino, Rodolfo
Tiny Neural Networks for Session-Level Traffic Classification Technical Report
2025.
@techreport{chehade2025tinyneuralnetworkssessionlevel,
title = {Tiny Neural Networks for Session-Level Traffic Classification},
author = {Adel Chehade and Edoardo Ragusa and Paolo Gastaldo and Rodolfo Zunino},
url = {https://arxiv.org/abs/2504.04008},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
3831.
Yan, Li; Du, Yi; Liang, Jing; Qu, Boyang; Li, Chao; Yu, Kunjie
A Weight Inheritance and Guidance Strategy based Evolutionary Network Architecture Search Journal Article
In: IEEE Transactions on Evolutionary Computation, pp. 1-1, 2025.
@article{10950438,
title = {A Weight Inheritance and Guidance Strategy based Evolutionary Network Architecture Search},
author = {Li Yan and Yi Du and Jing Liang and Boyang Qu and Chao Li and Kunjie Yu},
url = {https://ieeexplore.ieee.org/abstract/document/10950438},
doi = {10.1109/TEVC.2025.3558334},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Evolutionary Computation},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3830.
WAKAYAMA, Keigo; KANAMORI, Takafumi
Ultra-fast NAS based on Normalized Generalization Error with Random NTK Journal Article
In: IEICE Transactions on Information and Systems, vol. advpub, pp. 2024EDP7245, 2025.
@article{KeigoWAKAYAMA20252024EDP7245,
title = {Ultra-fast NAS based on Normalized Generalization Error with Random NTK},
author = {Keigo WAKAYAMA and Takafumi KANAMORI},
url = {https://www.jstage.jst.go.jp/article/transinf/advpub/0/advpub_2024EDP7245/_article},
doi = {10.1587/transinf.2024EDP7245},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEICE Transactions on Information and Systems},
volume = {advpub},
pages = {2024EDP7245},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
2024
3829.
Souza, Melwin D.; Prabhu, G. Ananth; Kumara, Varuna
Advanced Breast Cancer Detection Using Spatial Attention and Neural Architecture Search (SANAS-Net) Journal Article
In: SN Computer Science , vol. 6, 2024.
@article{nokey,
title = {Advanced Breast Cancer Detection Using Spatial Attention and Neural Architecture Search (SANAS-Net)},
author = {
Melwin D. Souza and G. Ananth Prabhu and Varuna Kumara
},
url = {https://link.springer.com/article/10.1007/s42979-024-03568-9},
year = {2024},
date = {2024-12-20},
urldate = {2024-12-20},
journal = {SN Computer Science },
volume = {6},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3828.
Alaoui, Ali Omari; Boutahir, Mohamed Khalifa; Bahi, Omaima El; Hessane, Abdelaaziz; Farhaoui, Yousef; Allaoui, Ahmad El
Accelerating deep learning model development—towards scalable automated architecture generation for optimal model design Journal Article
In: Multimedia Tools and Applications , 2024.
@article{Alaoui-mta24a,
title = {Accelerating deep learning model development—towards scalable automated architecture generation for optimal model design},
author = {Ali Omari Alaoui and Mohamed Khalifa Boutahir and Omaima El Bahi and Abdelaaziz Hessane and Yousef Farhaoui and Ahmad El Allaoui },
url = {https://link.springer.com/article/10.1007/s11042-024-20481-8},
year = {2024},
date = {2024-12-16},
urldate = {2024-12-16},
journal = {Multimedia Tools and Applications },
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3827.
Wang, Ruochen
Automated Machine Learning in the Era of Large Foundation Models PhD Thesis
2024.
@phdthesis{nokey,
title = {Automated Machine Learning in the Era of Large Foundation Models},
author = {Wang, Ruochen},
url = {https://escholarship.org/content/qt1vc4421f/qt1vc4421f.pdf},
year = {2024},
date = {2024-12-16},
urldate = {2024-12-16},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
3826.
Tan, Zhiwen; Guo, Daqi; Chen, Junan; Chen, Lei
M2M-Net: multi-objective neural architecture search using dynamic M2M population decomposition Journal Article
In: Neural Computing and Applications, 2024.
@article{tan-nca24a,
title = {M2M-Net: multi-objective neural architecture search using dynamic M2M population decomposition},
author = {
Zhiwen Tan and Daqi Guo and Junan Chen and Lei Chen
},
url = {https://link.springer.com/article/10.1007/s00521-024-10595-3},
year = {2024},
date = {2024-12-02},
urldate = {2024-12-02},
journal = { Neural Computing and Applications},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3825.
Zhou, Ao; Yang, Jianlei; Qi, Yingjie; Qiao, Tong; Shi, Yumeng; Duan, Cenlin; Zhao, Weisheng; Hu, Chunming
HGNAS: Hardware-Aware Graph Neural Architecture Search for Edge Devices Journal Article
In: IEEE Transactions on Computers, vol. 73, no. 12, pp. 2693-2707, 2024, ISSN: 1557-9956.
@article{10644077,
title = { HGNAS: Hardware-Aware Graph Neural Architecture Search for Edge Devices },
author = {Ao Zhou and Jianlei Yang and Yingjie Qi and Tong Qiao and Yumeng Shi and Cenlin Duan and Weisheng Zhao and Chunming Hu},
url = {https://doi.ieeecomputersociety.org/10.1109/TC.2024.3449108},
doi = {10.1109/TC.2024.3449108},
issn = {1557-9956},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
journal = {IEEE Transactions on Computers},
volume = {73},
number = {12},
pages = {2693-2707},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
abstract = {Graph Neural Networks (GNNs) are becoming increasingly popular for graph-based learning tasks such as point cloud processing due to their state-of-the-art (SOTA) performance. Nevertheless, the research community has primarily focused on improving model expressiveness, lacking consideration of how to design efficient GNN models for edge scenarios with real-time requirements and limited resources. Examining existing GNN models reveals varied execution across platforms and frequent Out-Of-Memory (OOM) problems, highlighting the need for hardware-aware GNN design. To address this challenge, this work proposes a novel hardware-aware graph neural architecture search framework tailored for resource constraint edge devices, namely HGNAS. To achieve hardware awareness, HGNAS integrates an efficient GNN hardware performance predictor that evaluates the latency and peak memory usage of GNNs in milliseconds. Meanwhile, we study GNN memory usage during inference and offer a peak memory estimation method, enhancing the robustness of architecture evaluations when combined with predictor outcomes. Furthermore, HGNAS constructs a fine-grained design space to enable the exploration of extreme performance architectures by decoupling the GNN paradigm. In addition, the multi-stage hierarchical search strategy is leveraged to facilitate the navigation of huge candidates, which can reduce the single search time to a few GPU hours. To the best of our knowledge, HGNAS is the first automated GNN design framework for edge devices, and also the first work to achieve hardware awareness of GNNs across different platforms. Extensive experiments across various applications and edge devices have proven the superiority of HGNAS. It can achieve up to a $10.6boldsymboltimes$10.6× speedup and an $82.5%$82.5% peak memory reduction with negligible accuracy loss compared to DGCNN on ModelNet40.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Graph Neural Networks (GNNs) are becoming increasingly popular for graph-based learning tasks such as point cloud processing due to their state-of-the-art (SOTA) performance. Nevertheless, the research community has primarily focused on improving model expressiveness, lacking consideration of how to design efficient GNN models for edge scenarios with real-time requirements and limited resources. Examining existing GNN models reveals varied execution across platforms and frequent Out-Of-Memory (OOM) problems, highlighting the need for hardware-aware GNN design. To address this challenge, this work proposes a novel hardware-aware graph neural architecture search framework tailored for resource constraint edge devices, namely HGNAS. To achieve hardware awareness, HGNAS integrates an efficient GNN hardware performance predictor that evaluates the latency and peak memory usage of GNNs in milliseconds. Meanwhile, we study GNN memory usage during inference and offer a peak memory estimation method, enhancing the robustness of architecture evaluations when combined with predictor outcomes. Furthermore, HGNAS constructs a fine-grained design space to enable the exploration of extreme performance architectures by decoupling the GNN paradigm. In addition, the multi-stage hierarchical search strategy is leveraged to facilitate the navigation of huge candidates, which can reduce the single search time to a few GPU hours. To the best of our knowledge, HGNAS is the first automated GNN design framework for edge devices, and also the first work to achieve hardware awareness of GNNs across different platforms. Extensive experiments across various applications and edge devices have proven the superiority of HGNAS. It can achieve up to a $10.6boldsymboltimes$10.6× speedup and an $82.5%$82.5% peak memory reduction with negligible accuracy loss compared to DGCNN on ModelNet40.
3824.
Shi, Xinjie; Guo, Chenxia; Yang, Ruifeng; Song, Yizhe
Adaptive evolutionary neural architecture search based on one-dimensional convolutional neural network for electric rudder fault diagnosis Journal Article
In: Measurement Science and Technology, vol. 36, no. 1, pp. 016038, 2024.
@article{Shi_2025,
title = {Adaptive evolutionary neural architecture search based on one-dimensional convolutional neural network for electric rudder fault diagnosis},
author = {Xinjie Shi and Chenxia Guo and Ruifeng Yang and Yizhe Song},
url = {https://dx.doi.org/10.1088/1361-6501/ad962e},
doi = {10.1088/1361-6501/ad962e},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
journal = {Measurement Science and Technology},
volume = {36},
number = {1},
pages = {016038},
publisher = {IOP Publishing},
abstract = {The electric rudder is the core actuator of the flight control system. Fault diagnosis of rudders is essential for the production and repair of rudders. While existing methods for rudder fault diagnosis are effective, the manual design of neural network models is a time-consuming and challenging process. Therefore, this paper proposes a fault diagnosis framework for the electric rudder based on an adaptive evolutionary neural architecture search (AENAS-FD). AENAS-FD employs an adaptive strategy to guide the evolution of a one-dimensional convolutional neural network towards achieving optimal diagnostic accuracy. This adaptive strategy adjusts the relevant parameters of the genetic operator based on the relationship between individual and population fitness. This leads to improved algorithm search performance and mitigates premature convergence. The experiments on the real electric rudder dataset demonstrate that AENAS-FD can generate superior network architectures for diagnosing rudder faults, exhibiting better diagnostic accuracy when compared to manually designed networks.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The electric rudder is the core actuator of the flight control system. Fault diagnosis of rudders is essential for the production and repair of rudders. While existing methods for rudder fault diagnosis are effective, the manual design of neural network models is a time-consuming and challenging process. Therefore, this paper proposes a fault diagnosis framework for the electric rudder based on an adaptive evolutionary neural architecture search (AENAS-FD). AENAS-FD employs an adaptive strategy to guide the evolution of a one-dimensional convolutional neural network towards achieving optimal diagnostic accuracy. This adaptive strategy adjusts the relevant parameters of the genetic operator based on the relationship between individual and population fitness. This leads to improved algorithm search performance and mitigates premature convergence. The experiments on the real electric rudder dataset demonstrate that AENAS-FD can generate superior network architectures for diagnosing rudder faults, exhibiting better diagnostic accuracy when compared to manually designed networks.
3823.
Sombo, Bem; Apeh, Simon Tooswem; Edeoghon, Isi Arthur
An Artificial Neural Network-Based Caller Authentication and Identification Algorithm in Cellular Communication Networks Journal Article
In: International Journal of Applied Methods in Electronics and Computers, 2024.
@article{nokey,
title = {An Artificial Neural Network-Based Caller Authentication and Identification Algorithm in Cellular Communication Networks},
author = { Bem Sombo and Simon Tooswem Apeh and Isi Arthur Edeoghon },
url = {https://www.ijamec.org/index.php/ijamec/article/view/432},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
journal = { International Journal of Applied Methods in Electronics and Computers},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3822.
Fouchal, Amel; Tikhamarine, Yazid; Benbouras, Mohammed Amin; Souag-Gamane, Doudja; Heddam, Salim
In: Modeling Earth Systems and Environment , vol. 11, 2024.
@article{Fouchal-mese24a,
title = {Biological oxygen demand prediction using artificial neural network and random forest models enhanced by the neural architecture search algorithm},
author = {Amel Fouchal and Yazid Tikhamarine and Mohammed Amin Benbouras and Doudja Souag-Gamane and Salim Heddam },
url = {https://link.springer.com/article/10.1007/s40808-024-02178-x},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
journal = {Modeling Earth Systems and Environment },
volume = {11},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3821.
Cheng, Jian; Jiang, Jinbo; Kang, Haidong; Ma, Lianbo
A Hybrid Neural Architecture Search Algorithm Optimized via Lifespan-PSO for Coal Mine Image Recognition Journal Article
In: Preprints, 2024.
@article{202412.2176,
title = {A Hybrid Neural Architecture Search Algorithm Optimized via Lifespan-PSO for Coal Mine Image Recognition},
author = {Jian Cheng and Jinbo Jiang and Haidong Kang and Lianbo Ma},
url = {https://doi.org/10.20944/preprints202412.2176.v1},
doi = {10.20944/preprints202412.2176.v1},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
journal = {Preprints},
publisher = {Preprints},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3820.
Garc'ıa, Jesús L. Llano; Monroy, Raúl; Hernández, V'ıctor Adrián Sosa
Neural architecture search for image super-resolution: A review on the emerging state-of-the-art Journal Article
In: Neurocomput., vol. 610, no. C, 2024, ISSN: 0925-2312.
@article{10.1016/j.neucom.2024.128481,
title = {Neural architecture search for image super-resolution: A review on the emerging state-of-the-art},
author = {Jesús L. Llano Garc'ıa and Raúl Monroy and V'ıctor Adrián Sosa Hernández},
url = {https://doi.org/10.1016/j.neucom.2024.128481},
doi = {10.1016/j.neucom.2024.128481},
issn = {0925-2312},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
journal = {Neurocomput.},
volume = {610},
number = {C},
publisher = {Elsevier Science Publishers B. V.},
address = {NLD},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3819.
AL-Sabri, Raeed; Gao, Jianliang; Chen, Jiamin; Oloulade, Babatounde Moctard; Wu, Zhenpeng; Abdullah, Monir; Hu, Xiaohua
M3GNAS: Multi-modal Multi-view Graph Neural Architecture Search for Medical Outcome Predictions Proceedings Article
In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pp. 1783-1788, IEEE Computer Society, Los Alamitos, CA, USA, 2024.
@inproceedings{10821927,
title = { M3GNAS: Multi-modal Multi-view Graph Neural Architecture Search for Medical Outcome Predictions },
author = {Raeed AL-Sabri and Jianliang Gao and Jiamin Chen and Babatounde Moctard Oloulade and Zhenpeng Wu and Monir Abdullah and Xiaohua Hu},
url = {https://doi.ieeecomputersociety.org/10.1109/BIBM62325.2024.10821927},
doi = {10.1109/BIBM62325.2024.10821927},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
booktitle = {2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)},
pages = {1783-1788},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
abstract = {Multi-modal multi-view graph learning models have achieved significant success in medical outcome prediction, combining various modalities to enhance the performance of various medical tasks. However, current architectures for multi-modal multi-view graph learning (M3GL) models heavily depend on manual design, demanding significant effort and expert experience. Meanwhile, significant advancements have been achieved in the field of graph neural architecture search (GNAS), contributing to the automated design of learning architectures based on graphs. However, GNAS faces challenges in automating multimodal multi-view graph learning (M3GL) models, as existing frameworks cannot handle M3GL architecture topology, and current search spaces do not consider M3GL models. To address the above challenges, we propose, for the first time, a multi-modal multi-view graph neural architecture search (M3GNAS) framework that automates the construction of the optimal M3GL models, enabling the integration of multi-modal features from different views. We also design an effective multi-modal multi-view learning (M3L) search space to develop inner-view and outer-view graph representation learning in the context of graph learning, obtaining a latent graph representation tailored to the specific requirements of downstream tasks. To examine the effectiveness of M3GNAS, it is evaluated on medical outcome prediction tasks. The experimental findings demonstrate our proposed framework’s superior performance compared to state-of-the-art models.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Multi-modal multi-view graph learning models have achieved significant success in medical outcome prediction, combining various modalities to enhance the performance of various medical tasks. However, current architectures for multi-modal multi-view graph learning (M3GL) models heavily depend on manual design, demanding significant effort and expert experience. Meanwhile, significant advancements have been achieved in the field of graph neural architecture search (GNAS), contributing to the automated design of learning architectures based on graphs. However, GNAS faces challenges in automating multimodal multi-view graph learning (M3GL) models, as existing frameworks cannot handle M3GL architecture topology, and current search spaces do not consider M3GL models. To address the above challenges, we propose, for the first time, a multi-modal multi-view graph neural architecture search (M3GNAS) framework that automates the construction of the optimal M3GL models, enabling the integration of multi-modal features from different views. We also design an effective multi-modal multi-view learning (M3L) search space to develop inner-view and outer-view graph representation learning in the context of graph learning, obtaining a latent graph representation tailored to the specific requirements of downstream tasks. To examine the effectiveness of M3GNAS, it is evaluated on medical outcome prediction tasks. The experimental findings demonstrate our proposed framework’s superior performance compared to state-of-the-art models.
3818.
Wu, Zhenpeng; Chen, Jiamin; Gao, Jianliang
Adaptive Drug Repositioning Prediction Proceedings Article
In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pp. 767-772, IEEE Computer Society, Los Alamitos, CA, USA, 2024.
@inproceedings{10822495,
title = { Adaptive Drug Repositioning Prediction },
author = {Zhenpeng Wu and Jiamin Chen and Jianliang Gao},
url = {https://doi.ieeecomputersociety.org/10.1109/BIBM62325.2024.10822495},
doi = {10.1109/BIBM62325.2024.10822495},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
booktitle = {2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)},
pages = {767-772},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
abstract = {Drug repositioning has become attractive because it can significantly accelerate drug discovery and reduce development costs by identifying new indications for existing drugs. Based on graph neural networks (GNNs), the recent work utilizes inter-domain information (drug-disease association network) and intra-domain information (drug-drug similarity network and disease-disease similarity network) to learn the effective representation of drugs and diseases. However, they overlook the significant impact of the adaptive inter-domain and intra-domain information fusion operation on model performance under different data-splitting strategies. Moreover, manually designing GNN architectures for specific drug repositioning datasets is time-consuming and expert-dependent. To address the above problem, we propose an adaptive drug repositioning prediction method called AdaDR, which can adaptively fuse inter-domain and intra-domain information under different data-splitting strategies and automatically design the optimal GNN architecture for each drug repositioning dataset. Specifically, we first design a unified drug repositioning search space with different information fusion operations and various handcrafted GNN architectures. Then, a drug repositioning model search will be adopted to enable an efficient search. Empirical studies on three benchmark datasets demonstrate that the optimal drug repositioning model identified by our proposed AdaDR achieves the best performance among competitive baselines. Through the analysis of the case study, the applicability of AdaDR in practical scenarios is further validated. The code is available at: https://github.com/csubigdata-Organization/AdaDR.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Drug repositioning has become attractive because it can significantly accelerate drug discovery and reduce development costs by identifying new indications for existing drugs. Based on graph neural networks (GNNs), the recent work utilizes inter-domain information (drug-disease association network) and intra-domain information (drug-drug similarity network and disease-disease similarity network) to learn the effective representation of drugs and diseases. However, they overlook the significant impact of the adaptive inter-domain and intra-domain information fusion operation on model performance under different data-splitting strategies. Moreover, manually designing GNN architectures for specific drug repositioning datasets is time-consuming and expert-dependent. To address the above problem, we propose an adaptive drug repositioning prediction method called AdaDR, which can adaptively fuse inter-domain and intra-domain information under different data-splitting strategies and automatically design the optimal GNN architecture for each drug repositioning dataset. Specifically, we first design a unified drug repositioning search space with different information fusion operations and various handcrafted GNN architectures. Then, a drug repositioning model search will be adopted to enable an efficient search. Empirical studies on three benchmark datasets demonstrate that the optimal drug repositioning model identified by our proposed AdaDR achieves the best performance among competitive baselines. Through the analysis of the case study, the applicability of AdaDR in practical scenarios is further validated. The code is available at: https://github.com/csubigdata-Organization/AdaDR.
3817.
Li, Gai; Cao, Chunhong; Fu, Huawei; Li, Xingxing; Gao, Xieping
Modeling Functional Brain Networks for ADHD via Spatial Preservation-Based Neural Architecture Search Journal Article
In: IEEE J Biomed Health Inform , 2024.
@article{Li-BHI24a,
title = { Modeling Functional Brain Networks for ADHD via Spatial Preservation-Based Neural Architecture Search },
author = {Gai Li and Chunhong Cao and Huawei Fu and Xingxing Li and Xieping Gao
},
url = {https://pubmed.ncbi.nlm.nih.gov/39167518/},
year = {2024},
date = {2024-11-28},
urldate = {2024-11-28},
journal = { IEEE J Biomed Health Inform },
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3816.
Min, Yongzhi; Jing, Qinglong; Li, Yaxing
Method for rail surface defect detection based on neural network architecture search Journal Article
In: Measurement Science and Technology, 2024.
@article{min-mst24a,
title = {Method for rail surface defect detection based on neural network architecture search},
author = { Yongzhi Min and Qinglong Jing and Yaxing Li },
url = {https://iopscience.iop.org/article/10.1088/1361-6501/ad9048},
doi = {10.1088/1361-6501/ad9048},
year = {2024},
date = {2024-11-20},
urldate = {2024-11-20},
journal = { Measurement Science and Technology},
keywords = {},
pubstate = {published},
tppubtype = {article}
}