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
4214.
Saghand, Esmat Ghasemi; Lai-Yuen, Susana K.
MONAS-ESNN: Multi-Objective Neural Architecture Search for Efficient Spiking Neural Networks Proceedings Article
In: 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 2963-2972, 2025.
@inproceedings{10944107,
title = {MONAS-ESNN: Multi-Objective Neural Architecture Search for Efficient Spiking Neural Networks},
author = {Esmat Ghasemi Saghand and Susana K. Lai-Yuen},
doi = {10.1109/WACV61041.2025.00891},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
pages = {2963-2972},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
4213.
Huang, Xinyuan; Gao, Jiechao
Federated Neural Architecture Search with Model-Agnostic Meta Learning Technical Report
2025.
@techreport{huang2025federatedneuralarchitecturesearch,
title = {Federated Neural Architecture Search with Model-Agnostic Meta Learning},
author = {Xinyuan Huang and Jiechao Gao},
url = {https://arxiv.org/abs/2504.06457},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4212.
Mohasel, Seyed Mojtaba; Sheppard, John; Molina, Lindsey K.; Neptune, Richard R.; Wurdeman, Shane R.; Pew, Corey A.
MicroNAS: An Automated Framework for Developing a Fall Detection System Technical Report
2025.
@techreport{mohasel2025micronasautomatedframeworkdeveloping,
title = {MicroNAS: An Automated Framework for Developing a Fall Detection System},
author = {Seyed Mojtaba Mohasel and John Sheppard and Lindsey K. Molina and Richard R. Neptune and Shane R. Wurdeman and Corey A. Pew},
url = {https://arxiv.org/abs/2504.07397},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4211.
Yang, Haitao; Liu, Zhaowei; Yang, Dong; Wang, Lihong
Parallel graph neural architecture search optimization with incomplete features Journal Article
In: Applied Soft Computing, vol. 176, pp. 113068, 2025, ISSN: 1568-4946.
@article{YANG2025113068,
title = {Parallel graph neural architecture search optimization with incomplete features},
author = {Haitao Yang and Zhaowei Liu and Dong Yang and Lihong Wang},
url = {https://www.sciencedirect.com/science/article/pii/S1568494625003795},
doi = {https://doi.org/10.1016/j.asoc.2025.113068},
issn = {1568-4946},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Applied Soft Computing},
volume = {176},
pages = {113068},
abstract = {Graph neural networks (GNNs) have shown remarkable success in many fields. However, the results of different model architectures for different scenarios can be very different. Designing effective neural architectures requires a great deal of specialized knowledge, which limits the application of GNNs models. In recent years, graph neural architecture search (GNAS) has attracted widespread attention. GNAS selects the GNNs structure in predefined search space using a suitable search algorithm. The search direction is constrained based on the evaluation made by the estimation strategy. Traditional GNAS methods suffer from long search times, difficulty in parameter selection, and high sensitivity to data quality. When feature information is missing, the candidate architectures explored during the search process cannot obtain complete feature information, which significantly reduces the accuracy of GNAS. To tackle these challenges, we propose a novel optimization framework for parallel graph neural architecture search, named AutoPGO. In AutoPGO, we complement the features based on a feature propagation algorithm generated by minimizing the Dirichlet energy function, improve the search algorithm using the mutation decay strategy and complete the optimization of the parameters using the Bayesian optimization method. Experimental results show that AutoPGO has good performance and some degree of robustness.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Graph neural networks (GNNs) have shown remarkable success in many fields. However, the results of different model architectures for different scenarios can be very different. Designing effective neural architectures requires a great deal of specialized knowledge, which limits the application of GNNs models. In recent years, graph neural architecture search (GNAS) has attracted widespread attention. GNAS selects the GNNs structure in predefined search space using a suitable search algorithm. The search direction is constrained based on the evaluation made by the estimation strategy. Traditional GNAS methods suffer from long search times, difficulty in parameter selection, and high sensitivity to data quality. When feature information is missing, the candidate architectures explored during the search process cannot obtain complete feature information, which significantly reduces the accuracy of GNAS. To tackle these challenges, we propose a novel optimization framework for parallel graph neural architecture search, named AutoPGO. In AutoPGO, we complement the features based on a feature propagation algorithm generated by minimizing the Dirichlet energy function, improve the search algorithm using the mutation decay strategy and complete the optimization of the parameters using the Bayesian optimization method. Experimental results show that AutoPGO has good performance and some degree of robustness.
4210.
Zhang, Kexin; Sun, Hao; Wei, Lixin; Hu, Ziyu
Dual-archive guided multi-objective neural architecture search with decomposition Journal Article
In: Expert Systems with Applications, vol. 282, pp. 127587, 2025, ISSN: 0957-4174.
@article{ZHANG2025127587,
title = {Dual-archive guided multi-objective neural architecture search with decomposition},
author = {Kexin Zhang and Hao Sun and Lixin Wei and Ziyu Hu},
url = {https://www.sciencedirect.com/science/article/pii/S0957417425012096},
doi = {https://doi.org/10.1016/j.eswa.2025.127587},
issn = {0957-4174},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Expert Systems with Applications},
volume = {282},
pages = {127587},
abstract = {The rapid development of generative artificial intelligence puts forward higher requirements for efficient neural network architecture design. Traditional neural architecture search (NAS) focuses on single-objective optimization, while multi-objective NAS (MONAS) needs to balance performance, parameter number, inference delay and other objectives at the same time. The existing methods based on multi-objective evolutionary algorithms often deviate from the optimal search direction due to target conflict, and rely on time-consuming weight pre-training, which seriously limits the practical application efficiency. Therefore, this paper proposes a dual-archive guided decomposition-based DANAS algorithm. The algorithm employs a dynamic decomposition strategy to map the multi-objective space into a set of subproblems. A convergence archive is utilized to preserve solutions approximating the Pareto front, while a diversity archive maintains the distribution characteristics of the solution set. The dual-archive mechanism effectively coordinates exploitation and exploration during the search process. On this basis, two training-free metrics are introduced, and an efficient variant without parameter training called TF-DANAS is proposed, which greatly reduces the search cost. In EvoXbench, NAS-Bench-101,and NAS-Bench-201 benchmark tests, DANAS algorithm obtains the optimal HV value on 57.6% of the test sets on EvoXbench. On NAS-Bench-101, the proposed method achieved a top average accuracy of 93.83% on CIFAR-10. Furthermore, it attained average accuracies of 94.34%, 73.42%, and 46.50% on CIFAR-10, CIFAR-100, and ImageNet16-120 respectively when evaluated on NAS-Bench-201.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The rapid development of generative artificial intelligence puts forward higher requirements for efficient neural network architecture design. Traditional neural architecture search (NAS) focuses on single-objective optimization, while multi-objective NAS (MONAS) needs to balance performance, parameter number, inference delay and other objectives at the same time. The existing methods based on multi-objective evolutionary algorithms often deviate from the optimal search direction due to target conflict, and rely on time-consuming weight pre-training, which seriously limits the practical application efficiency. Therefore, this paper proposes a dual-archive guided decomposition-based DANAS algorithm. The algorithm employs a dynamic decomposition strategy to map the multi-objective space into a set of subproblems. A convergence archive is utilized to preserve solutions approximating the Pareto front, while a diversity archive maintains the distribution characteristics of the solution set. The dual-archive mechanism effectively coordinates exploitation and exploration during the search process. On this basis, two training-free metrics are introduced, and an efficient variant without parameter training called TF-DANAS is proposed, which greatly reduces the search cost. In EvoXbench, NAS-Bench-101,and NAS-Bench-201 benchmark tests, DANAS algorithm obtains the optimal HV value on 57.6% of the test sets on EvoXbench. On NAS-Bench-101, the proposed method achieved a top average accuracy of 93.83% on CIFAR-10. Furthermore, it attained average accuracies of 94.34%, 73.42%, and 46.50% on CIFAR-10, CIFAR-100, and ImageNet16-120 respectively when evaluated on NAS-Bench-201.
4209.
Fang, Hui; Gao, Yang; Zhang, Peng; Yao, Jiangchao; Chen, Hongyang; Wang, Haishuai
Large Language Models Enhanced Personalized Graph Neural Architecture Search in Federated Learning Proceedings Article
In: Walsh, Toby; Shah, Julie; Kolter, Zico (Ed.): AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA, pp. 16514–16522, AAAI Press, 2025.
@inproceedings{DBLP:conf/aaai/Fang0ZYCW25,
title = {Large Language Models Enhanced Personalized Graph Neural Architecture Search in Federated Learning},
author = {Hui Fang and Yang Gao and Peng Zhang and Jiangchao Yao and Hongyang Chen and Haishuai Wang},
editor = {Toby Walsh and Julie Shah and Zico Kolter},
url = {https://doi.org/10.1609/aaai.v39i16.33814},
doi = {10.1609/AAAI.V39I16.33814},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {AAAI-25, Sponsored by the Association for the Advancement of Artificial
Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA},
pages = {16514–16522},
publisher = {AAAI Press},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
4208.
Wen, Yu; Zhang, Chen; Xie, Chenhao; Fu, Xin
Achieving Lightweight Super-Resolution for Real-Time Computer Graphics Proceedings Article
In: Walsh, Toby; Shah, Julie; Kolter, Zico (Ed.): AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA, pp. 8313–8322, AAAI Press, 2025.
@inproceedings{DBLP:conf/aaai/0003Z0F25,
title = {Achieving Lightweight Super-Resolution for Real-Time Computer Graphics},
author = {Yu Wen and Chen Zhang and Chenhao Xie and Xin Fu},
editor = {Toby Walsh and Julie Shah and Zico Kolter},
url = {https://doi.org/10.1609/aaai.v39i8.32897},
doi = {10.1609/AAAI.V39I8.32897},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {AAAI-25, Sponsored by the Association for the Advancement of Artificial
Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA},
pages = {8313–8322},
publisher = {AAAI Press},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
4207.
Maryam, Zara; VII, Researcher
Multi-Objective Evolutionary Optimization of Convolutional Neural Architectures Using Genetic Algorithms for Autonomous Visual Navigation Journal Article
In: vol. 6, pp. 8-13, 2025.
@article{articlei,
title = {Multi-Objective Evolutionary Optimization of Convolutional Neural Architectures Using Genetic Algorithms for Autonomous Visual Navigation},
author = {Zara Maryam and Researcher VII},
url = {https://www.researchgate.net/publication/390694093_Multi-Objective_Evolutionary_Optimization_of_Convolutional_Neural_Architectures_Using_Genetic_Algorithms_for_Autonomous_Visual_Navigation/citation/download},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
volume = {6},
pages = {8-13},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4206.
La, Hoang-Loc; Ha, Phuong Hoai
Kernel-Level Energy-Efficient Neural Architecture Search for Tabular Dataset Technical Report
2025.
@techreport{la2025kernellevelenergyefficientneuralarchitecture,
title = {Kernel-Level Energy-Efficient Neural Architecture Search for Tabular Dataset},
author = {Hoang-Loc La and Phuong Hoai Ha},
url = {https://arxiv.org/abs/2504.08359},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4205.
Cao, Bin; Zhao, Xianrui; Lyu, Zhihan
Evolutionary Intrusion Detection Strategy under Zero Trust Architecture Journal Article
In: IEEE Journal on Selected Areas in Communications, pp. 1-1, 2025.
@article{10963878,
title = {Evolutionary Intrusion Detection Strategy under Zero Trust Architecture},
author = {Bin Cao and Xianrui Zhao and Zhihan Lyu},
url = {https://ieeexplore.ieee.org/abstract/document/10963878},
doi = {10.1109/JSAC.2025.3560001},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Journal on Selected Areas in Communications},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4204.
Xie, Shubing; Sarkar, Aritra; Feld, Sebastian
DeQompile: quantum circuit decompilation using genetic programming for explainable quantum architecture search Miscellaneous
2025.
@misc{xie2025deqompilequantumcircuitdecompilation,
title = {DeQompile: quantum circuit decompilation using genetic programming for explainable quantum architecture search},
author = {Shubing Xie and Aritra Sarkar and Sebastian Feld},
url = {https://arxiv.org/abs/2504.08310},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
4203.
Wang, Wei; Wang, Xianpeng; and, Zhiming Dong
Neural architecture search for microscopic image segmentation using a constrained multi-objective evolutionary algorithm Journal Article
In: Engineering Optimization, vol. 0, no. 0, pp. 1–20, 2025.
@article{Wang14042025,
title = {Neural architecture search for microscopic image segmentation using a constrained multi-objective evolutionary algorithm},
author = {Wei Wang and Xianpeng Wang and Zhiming Dong and},
url = {https://doi.org/10.1080/0305215X.2025.2464852},
doi = {10.1080/0305215X.2025.2464852},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Engineering Optimization},
volume = {0},
number = {0},
pages = {1–20},
publisher = {Taylor & Francis},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4202.
Papic, Admir; Herenda, Faruk; Hubana, Tarik; Hodzic, Migdat
Beyond Trial and Error: Comparative Analysis of Heuristic Strategies for Optimal Neural Network Architectures Proceedings Article
In: 2025 24th International Symposium INFOTEH-JAHORINA (INFOTEH), pp. 1-6, 2025.
@inproceedings{10959303,
title = {Beyond Trial and Error: Comparative Analysis of Heuristic Strategies for Optimal Neural Network Architectures},
author = {Admir Papic and Faruk Herenda and Tarik Hubana and Migdat Hodzic},
url = {https://ieeexplore.ieee.org/abstract/document/10959303},
doi = {10.1109/INFOTEH64129.2025.10959303},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {2025 24th International Symposium INFOTEH-JAHORINA (INFOTEH)},
pages = {1-6},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
4201.
Zeinaty, Christophe El; Hamidouche, Wassim; Herrou, Glenn; Menard, Daniel; Debbah, Merouane
Can LLMs Revolutionize the Design of Explainable and Efficient TinyML Models? Technical Report
2025.
@techreport{zeinaty2025llmsrevolutionizedesignexplainable,
title = {Can LLMs Revolutionize the Design of Explainable and Efficient TinyML Models?},
author = {Christophe El Zeinaty and Wassim Hamidouche and Glenn Herrou and Daniel Menard and Merouane Debbah},
url = {https://arxiv.org/abs/2504.09685},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4200.
Ebadi, Ali; Kaur, Manpreet; Liu, Qian
Hyperparameter optimization and neural architecture search algorithms for graph Neural Networks in cheminformatics Journal Article
In: Computational Materials Science, vol. 254, pp. 113904, 2025, ISSN: 0927-0256.
@article{EBADI2025113904,
title = {Hyperparameter optimization and neural architecture search algorithms for graph Neural Networks in cheminformatics},
author = {Ali Ebadi and Manpreet Kaur and Qian Liu},
url = {https://www.sciencedirect.com/science/article/pii/S0927025625002472},
doi = {https://doi.org/10.1016/j.commatsci.2025.113904},
issn = {0927-0256},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Computational Materials Science},
volume = {254},
pages = {113904},
abstract = {Cheminformatics, an interdisciplinary field bridging chemistry and information science, leverages computational tools to analyze and interpret chemical data, playing a critical role in drug discovery, material science, and environmental chemistry. Traditional methods, reliant on rule-based algorithms and expert-curated datasets, face challenges in scalability and adaptability. Recently, machine learning and deep learning have revolutionized cheminformatics by offering data-driven approaches that uncover complex patterns in vast chemical datasets, advancing molecular property prediction, chemical reaction modeling, and de novo molecular design. Among the most promising techniques are Graph Neural Networks (GNNs), which have emerged as a powerful tool for modeling molecules in a manner that mirrors their underlying chemical structures. Despite their success, the performance of GNNs is highly sensitive to architectural choices and hyperparameters, making optimal configuration selection a non-trivial task. Neural Architecture Search (NAS) and Hyperparameter Optimization (HPO) are crucial for improving GNN performance, but the complexity and computational cost of these processes have traditionally hindered progress. This review examines various strategies for automating NAS and HPO in GNNs, highlighting their potential to enhance model performance, scalability, and efficiency in key cheminformatics applications. As the field evolves, automated optimization techniques are expected to play a pivotal role in advancing GNN-based solutions in cheminformatics.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Cheminformatics, an interdisciplinary field bridging chemistry and information science, leverages computational tools to analyze and interpret chemical data, playing a critical role in drug discovery, material science, and environmental chemistry. Traditional methods, reliant on rule-based algorithms and expert-curated datasets, face challenges in scalability and adaptability. Recently, machine learning and deep learning have revolutionized cheminformatics by offering data-driven approaches that uncover complex patterns in vast chemical datasets, advancing molecular property prediction, chemical reaction modeling, and de novo molecular design. Among the most promising techniques are Graph Neural Networks (GNNs), which have emerged as a powerful tool for modeling molecules in a manner that mirrors their underlying chemical structures. Despite their success, the performance of GNNs is highly sensitive to architectural choices and hyperparameters, making optimal configuration selection a non-trivial task. Neural Architecture Search (NAS) and Hyperparameter Optimization (HPO) are crucial for improving GNN performance, but the complexity and computational cost of these processes have traditionally hindered progress. This review examines various strategies for automating NAS and HPO in GNNs, highlighting their potential to enhance model performance, scalability, and efficiency in key cheminformatics applications. As the field evolves, automated optimization techniques are expected to play a pivotal role in advancing GNN-based solutions in cheminformatics.
4199.
Li, Qing; Chen, Jingrun
An Unsupervised Network Architecture Search Method for Solving Partial Differential Equations Technical Report
2025.
@techreport{li2025unsupervisednetworkarchitecturesearch,
title = {An Unsupervised Network Architecture Search Method for Solving Partial Differential Equations},
author = {Qing Li and Jingrun Chen},
url = {https://arxiv.org/abs/2504.11140},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4198.
Litty, Abey; Okunola, Abiodun; Lima, Gabion
Automatically Discovering Novel and Efficient Algorithmic Structures Using Deep Learning Technical Report
2025.
@techreport{articlej,
title = {Automatically Discovering Novel and Efficient Algorithmic Structures Using Deep Learning},
author = {Abey Litty and Abiodun Okunola and Gabion Lima},
url = {https://www.researchgate.net/publication/390802620_Automatically_Discovering_Novel_and_Efficient_Algorithmic_Structures_Using_Deep_Learning/citations},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4197.
Ahsun, Abbas; Noah, Asher; John, Ada
Building intelligent systems that can automatically generate tailored algorithms for specific application areas Technical Report
2025.
@techreport{articlek,
title = {Building intelligent systems that can automatically generate tailored algorithms for specific application areas},
author = {Abbas Ahsun and Asher Noah and Ada John},
url = {https://www.researchgate.net/publication/390805745_Building_intelligent_systems_that_can_automatically_generate_tailored_algorithms_for_specific_application_areas},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4196.
Carstensen, Timur; Mallik, Neeratyoy; Hutter, Frank; Rapp, Martin
Frozen Layers: Memory-efficient Many-fidelity Hyperparameter Optimization Technical Report
2025.
@techreport{carstensen2025frozenlayersmemoryefficientmanyfidelity,
title = {Frozen Layers: Memory-efficient Many-fidelity Hyperparameter Optimization},
author = {Timur Carstensen and Neeratyoy Mallik and Frank Hutter and Martin Rapp},
url = {https://arxiv.org/abs/2504.10735},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4195.
Qin, Shiwen; Kadlecová, Gabriela; Pilát, Martin; Cohen, Shay B.; Neruda, Roman; Crowley, Elliot J.; Lukasik, Jovita; Ericsson, Linus
Transferrable Surrogates in Expressive Neural Architecture Search Spaces Technical Report
2025.
@techreport{qin2025transferrablesurrogatesexpressiveneural,
title = {Transferrable Surrogates in Expressive Neural Architecture Search Spaces},
author = {Shiwen Qin and Gabriela Kadlecová and Martin Pilát and Shay B. Cohen and Roman Neruda and Elliot J. Crowley and Jovita Lukasik and Linus Ericsson},
url = {https://arxiv.org/abs/2504.12971},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4194.
Liu, Zhi-Ang; Liu, Jiang-Jiang
Towards efficient salient object detection via U-shape architecture search Journal Article
In: Knowledge-Based Systems, vol. 318, pp. 113515, 2025, ISSN: 0950-7051.
@article{LIU2025113515,
title = {Towards efficient salient object detection via U-shape architecture search},
author = {Zhi-Ang Liu and Jiang-Jiang Liu},
url = {https://www.sciencedirect.com/science/article/pii/S0950705125005611},
doi = {https://doi.org/10.1016/j.knosys.2025.113515},
issn = {0950-7051},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Knowledge-Based Systems},
volume = {318},
pages = {113515},
abstract = {State-of-the-art (SOTA) deep learning architectures for salient object detection (SOD) are predominantly handcrafted, with many methods adapting successful classification backbones for SOD tasks. In this work, we aim to design more efficient architectures specifically tailored for SOD, accommodating both RGB and RGB-D input modalities. Leveraging neural architecture search (NAS), we uncover unique characteristics of SOD by proposing a novel U-shaped search space. This design integrates bottom-up and top-down pathways, enhanced by interconnecting links between them. To further optimize the complexity distribution across these pathways, we introduce a per-complexity importance measure. The resulting architectures, NASAL and NASAL-D (targeting RGB and RGB-D SOD, respectively), achieve an improved trade-off between accuracy and computational efficiency. Extensive experiments on multiple popular benchmarks demonstrate that our models consistently outperform existing methods in both effectiveness and efficiency.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
State-of-the-art (SOTA) deep learning architectures for salient object detection (SOD) are predominantly handcrafted, with many methods adapting successful classification backbones for SOD tasks. In this work, we aim to design more efficient architectures specifically tailored for SOD, accommodating both RGB and RGB-D input modalities. Leveraging neural architecture search (NAS), we uncover unique characteristics of SOD by proposing a novel U-shaped search space. This design integrates bottom-up and top-down pathways, enhanced by interconnecting links between them. To further optimize the complexity distribution across these pathways, we introduce a per-complexity importance measure. The resulting architectures, NASAL and NASAL-D (targeting RGB and RGB-D SOD, respectively), achieve an improved trade-off between accuracy and computational efficiency. Extensive experiments on multiple popular benchmarks demonstrate that our models consistently outperform existing methods in both effectiveness and efficiency.
4193.
Saha, Bidyut; Samanta, Riya; Roy, Ram Babu; Ghosh, Soumya K.
TinyTNAS: Time-Bound, GPU-Independent Hardware-Aware Neural Architecture Search for TinyML Time Series Classification Journal Article
In: IEEE Embedded Systems Letters, pp. 1-1, 2025.
@article{10967537,
title = {TinyTNAS: Time-Bound, GPU-Independent Hardware-Aware Neural Architecture Search for TinyML Time Series Classification},
author = {Bidyut Saha and Riya Samanta and Ram Babu Roy and Soumya K. Ghosh},
url = {https://ieeexplore.ieee.org/abstract/document/10967537},
doi = {10.1109/LES.2025.3561870},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Embedded Systems Letters},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4192.
Lu, Xiaotong; Dong, Weisheng; Fang, Zhenxuan; Lin, Jie; Li, Xin; Shi, Guangming
Growing-before-pruning: A progressive neural architecture search strategy via group sparsity and deterministic annealing Journal Article
In: Pattern Recognition, vol. 166, pp. 111697, 2025, ISSN: 0031-3203.
@article{LU2025111697,
title = {Growing-before-pruning: A progressive neural architecture search strategy via group sparsity and deterministic annealing},
author = {Xiaotong Lu and Weisheng Dong and Zhenxuan Fang and Jie Lin and Xin Li and Guangming Shi},
url = {https://www.sciencedirect.com/science/article/pii/S0031320325003577},
doi = {https://doi.org/10.1016/j.patcog.2025.111697},
issn = {0031-3203},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Pattern Recognition},
volume = {166},
pages = {111697},
abstract = {Network pruning is a widely studied technique of obtaining compact representations from over-parameterized deep convolutional neural networks. Existing pruning methods are based on finding an optimal combination of pruned filters in the fixed search space. However, the optimality of those methods is often questionable due to limited search space and pruning choices - e.g., the difficulty with removing the entire layer and the risk of unexpected performance degradation. Inspired by the exploration vs. exploitation trade-off in reinforcement learning, we propose to reconstruct the filter space without increasing the model capacity and prune them by exploiting group sparsity. Our approach challenges the conventional wisdom by advocating the strategy of Growing-before-Pruning (GbP), which allows us to explore more space before exploiting the power of architecture search. Meanwhile, to achieve more efficient pruning, we propose to measure the importance of filters by global group sparsity, which extends the existing Gaussian scale mixture model. Such global characterization of sparsity in the filter space leads to a novel deterministic annealing strategy for progressively pruning the filters. We have evaluated our method on several popular datasets and network architectures. Our extensive experiment results have shown that the proposed method advances the current state-of-the-art.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Network pruning is a widely studied technique of obtaining compact representations from over-parameterized deep convolutional neural networks. Existing pruning methods are based on finding an optimal combination of pruned filters in the fixed search space. However, the optimality of those methods is often questionable due to limited search space and pruning choices - e.g., the difficulty with removing the entire layer and the risk of unexpected performance degradation. Inspired by the exploration vs. exploitation trade-off in reinforcement learning, we propose to reconstruct the filter space without increasing the model capacity and prune them by exploiting group sparsity. Our approach challenges the conventional wisdom by advocating the strategy of Growing-before-Pruning (GbP), which allows us to explore more space before exploiting the power of architecture search. Meanwhile, to achieve more efficient pruning, we propose to measure the importance of filters by global group sparsity, which extends the existing Gaussian scale mixture model. Such global characterization of sparsity in the filter space leads to a novel deterministic annealing strategy for progressively pruning the filters. We have evaluated our method on several popular datasets and network architectures. Our extensive experiment results have shown that the proposed method advances the current state-of-the-art.
4191.
Mohasel, Seyed Mojtaba; Sheppard, John; Molina, Lindsey K.; Neptune, Richard R.; Wurdeman, Shane R.; Pew, Corey A.
MicroNAS: An Automated Framework for Developing a Fall Detection System Technical Report
2025.
@techreport{mohasel2025micronasautomatedframeworkdevelopingb,
title = {MicroNAS: An Automated Framework for Developing a Fall Detection System},
author = {Seyed Mojtaba Mohasel and John Sheppard and Lindsey K. Molina and Richard R. Neptune and Shane R. Wurdeman and Corey A. Pew},
url = {https://arxiv.org/abs/2504.07397},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4190.
Zhao, Yunbiao; Wang, Lei; Wu, Jiang; Li, Taiyong
NAST: neural architecture search and transformer for image manipulation localization Journal Article
In: Journal of Electronic Imaging, vol. 34, no. 2, pp. 023057, 2025.
@article{10.1117/1.JEI.34.2.023057,
title = {NAST: neural architecture search and transformer for image manipulation localization},
author = {Yunbiao Zhao and Lei Wang and Jiang Wu and Taiyong Li},
url = {https://doi.org/10.1117/1.JEI.34.2.023057},
doi = {10.1117/1.JEI.34.2.023057},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Journal of Electronic Imaging},
volume = {34},
number = {2},
pages = {023057},
publisher = {SPIE},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4189.
Zhang, Ruohan; Li, Lingling; Jiao, Licheng; Liu, Fang; Liu, Xu; Yang, Shuyuan
Knowledge-Aware Evolutionary Transformer Journal Article
In: IEEE Transactions on Evolutionary Computation, pp. 1-1, 2025.
@article{10970744,
title = {Knowledge-Aware Evolutionary Transformer},
author = {Ruohan Zhang and Lingling Li and Licheng Jiao and Fang Liu and Xu Liu and Shuyuan Yang},
url = {https://ieeexplore.ieee.org/document/10970744},
doi = {10.1109/TEVC.2025.3562576},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Evolutionary Computation},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4188.
Liu, Wenbo; Wu, Jia; Deng, Tao; Yan, Fei
Continuous–Discrete Alignment Optimization for efficient differentiable neural architecture search Journal Article
In: Engineering Applications of Artificial Intelligence, vol. 153, pp. 110721, 2025, ISSN: 0952-1976.
@article{LIU2025110721,
title = {Continuous–Discrete Alignment Optimization for efficient differentiable neural architecture search},
author = {Wenbo Liu and Jia Wu and Tao Deng and Fei Yan},
url = {https://www.sciencedirect.com/science/article/pii/S0952197625007213},
doi = {https://doi.org/10.1016/j.engappai.2025.110721},
issn = {0952-1976},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Engineering Applications of Artificial Intelligence},
volume = {153},
pages = {110721},
abstract = {Differential Architecture Search (DARTS) has become a prominent technique for neural architecture search in recent years. Despite its merits, the issue of discretization discrepancy within DARTS still necessitates further exploration, as it can degrade in performance. In this paper, we introduce a novel algorithm termed Continuous–Discrete Alignment Optimization (DARTS-CDAO), designed to address the discretization discrepancy and thereby enhance the robustness and generalization capabilities of the discovered neural architectures. Our proposed DARTS-CDAO algorithm seamlessly integrates the discretization process into the training phase of the architecture parameters, thereby bolstering the search algorithm’s adaptability to the inherent discretization processes. Specifically, our methodology commences by formalizing the process of architecture parameter discretization. Subsequently, we introduce a coarse gradient weighting algorithm that is employed to update the architecture parameters, effectively minimizing the divergence between the representation of continuous and discrete parameters. Rigorous theoretical analysis, coupled with extensive experimental outcomes, substantiates that our proposed approach can elevate the performance of the searched models. Notably, this enhancement is achieved without incurring additional search time, rendering DARTS more robust and endowed with a heightened capacity for generalization.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Differential Architecture Search (DARTS) has become a prominent technique for neural architecture search in recent years. Despite its merits, the issue of discretization discrepancy within DARTS still necessitates further exploration, as it can degrade in performance. In this paper, we introduce a novel algorithm termed Continuous–Discrete Alignment Optimization (DARTS-CDAO), designed to address the discretization discrepancy and thereby enhance the robustness and generalization capabilities of the discovered neural architectures. Our proposed DARTS-CDAO algorithm seamlessly integrates the discretization process into the training phase of the architecture parameters, thereby bolstering the search algorithm’s adaptability to the inherent discretization processes. Specifically, our methodology commences by formalizing the process of architecture parameter discretization. Subsequently, we introduce a coarse gradient weighting algorithm that is employed to update the architecture parameters, effectively minimizing the divergence between the representation of continuous and discrete parameters. Rigorous theoretical analysis, coupled with extensive experimental outcomes, substantiates that our proposed approach can elevate the performance of the searched models. Notably, this enhancement is achieved without incurring additional search time, rendering DARTS more robust and endowed with a heightened capacity for generalization.
4187.
Mecharbat, Lotfi Abdelkrim; Almakky, Ibrahim; Takac, Martin; Yaqub, Mohammad
MedNNS: Supernet-based Medical Task-Adaptive Neural Network Search Technical Report
2025.
@techreport{mecharbat2025mednnssupernetbasedmedicaltaskadaptive,
title = {MedNNS: Supernet-based Medical Task-Adaptive Neural Network Search},
author = {Lotfi Abdelkrim Mecharbat and Ibrahim Almakky and Martin Takac and Mohammad Yaqub},
url = {https://arxiv.org/abs/2504.15865},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4186.
Li, Kai; Huang, Mingqiang; Li, Ang; Yang, Shuxin; Cheng, Quan; Yu, Hao
A 29.12-TOPS/W Vector Systolic Accelerator With NAS-Optimized DNNs in 28-nm CMOS Journal Article
In: IEEE Journal of Solid-State Circuits, pp. 1-12, 2025.
@article{10972309,
title = {A 29.12-TOPS/W Vector Systolic Accelerator With NAS-Optimized DNNs in 28-nm CMOS},
author = {Kai Li and Mingqiang Huang and Ang Li and Shuxin Yang and Quan Cheng and Hao Yu},
url = {https://ieeexplore.ieee.org/abstract/document/10972309},
doi = {10.1109/JSSC.2025.3558287},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Journal of Solid-State Circuits},
pages = {1-12},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4185.
Xie, Cheng; Xue, Bing; Yang, Mingyi; Zhang, Mengjie; Xu, Zhigang; Wang, Junyi; Chen, Songlin; Liu, Yang
Evolutionary neural architecture search for automatically designing CNN-GRU-Attention neural networks for turntable servo systems Journal Article
In: Expert Systems with Applications, vol. 283, pp. 127765, 2025, ISSN: 0957-4174.
@article{XIE2025127765,
title = {Evolutionary neural architecture search for automatically designing CNN-GRU-Attention neural networks for turntable servo systems},
author = {Cheng Xie and Bing Xue and Mingyi Yang and Mengjie Zhang and Zhigang Xu and Junyi Wang and Songlin Chen and Yang Liu},
url = {https://www.sciencedirect.com/science/article/pii/S0957417425013879},
doi = {https://doi.org/10.1016/j.eswa.2025.127765},
issn = {0957-4174},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Expert Systems with Applications},
volume = {283},
pages = {127765},
abstract = {Turntable servo systems are important experimental devices utilized in the semi-physical simulation and testing of aircraft. Building a model for turntable servo systems, which can accurately predict their operating states or behaviors, is important for the development and debugging of the whole servo control system. However, servo systems usually have complex nonlinear characteristics and external disturbances, which brings challenges to the accurate modeling of them. Therefore, this paper develops a modeling method for turntable servo systems by using CNN-GRU-Attention hybrid neural network models. The CNN-GRU-Attention model is used for compensating the nonlinear terms of the known dynamics model, thus effectively improving the modeling accuracy. Considering the complex architectures of the above hybrid model, an evolutionary neural architecture search (ENAS) algorithm is proposed accordingly, which can automatically design the architecture of the CNN-GRU-Attention models. During the designing process, a variable-length encoding strategy is proposed to represent the possible architectures, and novel crossover and mutation operators are proposed accordingly for the evolution of individuals. In addition, the real-time constraints are considered in the design of the search space, so that all the searched models can meet the real-time requirements. The experimental results show the CNN-GRU-Attention models obtained by the proposed ENAS algorithm achieve superior prediction performance over the peer competitors in most tasks. The effectiveness of the proposed method is further verified by analyzing the convergence and the search results.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Turntable servo systems are important experimental devices utilized in the semi-physical simulation and testing of aircraft. Building a model for turntable servo systems, which can accurately predict their operating states or behaviors, is important for the development and debugging of the whole servo control system. However, servo systems usually have complex nonlinear characteristics and external disturbances, which brings challenges to the accurate modeling of them. Therefore, this paper develops a modeling method for turntable servo systems by using CNN-GRU-Attention hybrid neural network models. The CNN-GRU-Attention model is used for compensating the nonlinear terms of the known dynamics model, thus effectively improving the modeling accuracy. Considering the complex architectures of the above hybrid model, an evolutionary neural architecture search (ENAS) algorithm is proposed accordingly, which can automatically design the architecture of the CNN-GRU-Attention models. During the designing process, a variable-length encoding strategy is proposed to represent the possible architectures, and novel crossover and mutation operators are proposed accordingly for the evolution of individuals. In addition, the real-time constraints are considered in the design of the search space, so that all the searched models can meet the real-time requirements. The experimental results show the CNN-GRU-Attention models obtained by the proposed ENAS algorithm achieve superior prediction performance over the peer competitors in most tasks. The effectiveness of the proposed method is further verified by analyzing the convergence and the search results.
4184.
Zhou, Yanlin; El-Khamy, Mostafa; Song, Kee-Bong
Regularizing Differentiable Architecture Search with Smooth Activation Technical Report
2025.
@techreport{zhou2025regularizingdifferentiablearchitecturesearch,
title = {Regularizing Differentiable Architecture Search with Smooth Activation},
author = {Yanlin Zhou and Mostafa El-Khamy and Kee-Bong Song},
url = {https://arxiv.org/abs/2504.16306},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4183.
Liu, Xinyu; Xia, Ding; Yan, Siqi; Lv, Haoran; Wang, Aili; Wu, Haibin
Noise disruption inspired NAS with efficient focal attention for hyperspectral image classification Proceedings Article
In: Wu, Haibin (Ed.): International Conference on Measurement, Communication, and Virtual Reality (MCVR 2024), pp. 1363404, International Society for Optics and Photonics SPIE, 2025.
@inproceedings{10.1117/12.3066039,
title = {Noise disruption inspired NAS with efficient focal attention for hyperspectral image classification},
author = {Xinyu Liu and Ding Xia and Siqi Yan and Haoran Lv and Aili Wang and Haibin Wu},
editor = {Haibin Wu},
url = {https://doi.org/10.1117/12.3066039},
doi = {10.1117/12.3066039},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {International Conference on Measurement, Communication, and Virtual Reality (MCVR 2024)},
volume = {13634},
pages = {1363404},
publisher = {SPIE},
organization = {International Society for Optics and Photonics},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
4182.
Aslan, Bilal; Kazaka, Winner; Slaven, Tomas; Chetty, Shaylin; Kruger, Nicholas; Nitschke, Geoff
Deep-Learning Classifiers for Small Data Orthopedic Radiology Proceedings Article
In: 2025 IEEE Symposium on Computational Intelligence in Health and Medicine (CIHM), pp. 1-7, 2025.
@inproceedings{10969486,
title = {Deep-Learning Classifiers for Small Data Orthopedic Radiology},
author = {Bilal Aslan and Winner Kazaka and Tomas Slaven and Shaylin Chetty and Nicholas Kruger and Geoff Nitschke},
url = {https://ieeexplore.ieee.org/abstract/document/10969486},
doi = {10.1109/CIHM64979.2025.10969486},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {2025 IEEE Symposium on Computational Intelligence in Health and Medicine (CIHM)},
pages = {1-7},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
4181.
Xuejian, Zhao; Wenxin, Chen; Enliang, Wang; Yekai, Hu
DPSO-NAS: Wall Crack Detection Algorithm Based on Particle Swarm Optimization NAS Journal Article
In: IEEE Transactions on Consumer Electronics, pp. 1-1, 2025.
@article{10976256,
title = {DPSO-NAS: Wall Crack Detection Algorithm Based on Particle Swarm Optimization NAS},
author = {Zhao Xuejian and Chen Wenxin and Wang Enliang and Hu Yekai},
url = {https://ieeexplore.ieee.org/abstract/document/10976256},
doi = {10.1109/TCE.2025.3564011},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Consumer Electronics},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4180.
Tong, Guoxiang; Qian, Junjie; Shen, Jieyu
Adaptive metagraph neural network assisted by metagraph search for financial fraud detection Journal Article
In: Engineering Applications of Artificial Intelligence, vol. 153, pp. 110807, 2025, ISSN: 0952-1976.
@article{TONG2025110807,
title = {Adaptive metagraph neural network assisted by metagraph search for financial fraud detection},
author = {Guoxiang Tong and Junjie Qian and Jieyu Shen},
url = {https://www.sciencedirect.com/science/article/pii/S0952197625008073},
doi = {https://doi.org/10.1016/j.engappai.2025.110807},
issn = {0952-1976},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Engineering Applications of Artificial Intelligence},
volume = {153},
pages = {110807},
abstract = {Financial transaction fraud detection is a critical technology for ensuring the security and stability of financial markets. Artificial intelligence, particularly graph neural networks, has demonstrated superior performance in fraud detection. However, challenges remain, such as limited interpretability, difficulty in adapting to new types of fraud in a timely manner, and incomplete data mining. To address these challenges, we propose a novel graph neural network model called Metagraph Fraud Detection Graph Neural Networks (MetaFraud-GNN), which leverages metagraph search and neural architecture search (NAS) techniques to automatically optimize the network structure for financial transaction fraud detection. MetaFraud-GNN extracts complex patterns from financial transaction networks through metagraph search algorithms, a technique that automatically mines key subgraph patterns. These metagraph capture fraudulent patterns and enable the model to more comprehensively uncover hidden information within transaction data, thus enhancing its processing efficiency. Additionally, the metagraph decoding algorithm optimizes the graph neural network structure by training on the most effective metagraph to adapt to evolving fraudulent methods. This approach improves both the accuracy and adaptability of fraud detection. We conduct experiments on three real-world public benchmark datasets YelpChi, Amazon, and Elliptic and demonstrate that our model significantly outperforms existing benchmark methods on various performance metrics. Such as F1-macro, Area Under the Receiver Operating Characteristic Curve(AUC) and Geometric Mean(GMean), by 4.46%, 2.67%, and 8.59% on YelpChi, 0.14% and 2.19% on Amazon,The F1 indicator has not been upgraded, respectively.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Financial transaction fraud detection is a critical technology for ensuring the security and stability of financial markets. Artificial intelligence, particularly graph neural networks, has demonstrated superior performance in fraud detection. However, challenges remain, such as limited interpretability, difficulty in adapting to new types of fraud in a timely manner, and incomplete data mining. To address these challenges, we propose a novel graph neural network model called Metagraph Fraud Detection Graph Neural Networks (MetaFraud-GNN), which leverages metagraph search and neural architecture search (NAS) techniques to automatically optimize the network structure for financial transaction fraud detection. MetaFraud-GNN extracts complex patterns from financial transaction networks through metagraph search algorithms, a technique that automatically mines key subgraph patterns. These metagraph capture fraudulent patterns and enable the model to more comprehensively uncover hidden information within transaction data, thus enhancing its processing efficiency. Additionally, the metagraph decoding algorithm optimizes the graph neural network structure by training on the most effective metagraph to adapt to evolving fraudulent methods. This approach improves both the accuracy and adaptability of fraud detection. We conduct experiments on three real-world public benchmark datasets YelpChi, Amazon, and Elliptic and demonstrate that our model significantly outperforms existing benchmark methods on various performance metrics. Such as F1-macro, Area Under the Receiver Operating Characteristic Curve(AUC) and Geometric Mean(GMean), by 4.46%, 2.67%, and 8.59% on YelpChi, 0.14% and 2.19% on Amazon,The F1 indicator has not been upgraded, respectively.
4179.
Zhou, Bingye; Yu, Caiyang
Evolution Meets Diffusion: Efficient Neural Architecture Generation Technical Report
2025.
@techreport{zhou2025evolutionmeetsdiffusionefficient,
title = {Evolution Meets Diffusion: Efficient Neural Architecture Generation},
author = {Bingye Zhou and Caiyang Yu},
url = {https://arxiv.org/abs/2504.17827},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4178.
Louati, Hassen; Louati, Ali; Kariri, Elham; Almekhlafi, Abdulla
AI-Based Anomaly Detection and Optimization Framework for Blockchain Smart Contracts Journal Article
In: Administrative Sciences, vol. 15, no. 5, 2025, ISSN: 2076-3387.
@article{admsci15050163,
title = {AI-Based Anomaly Detection and Optimization Framework for Blockchain Smart Contracts},
author = {Hassen Louati and Ali Louati and Elham Kariri and Abdulla Almekhlafi},
url = {https://www.mdpi.com/2076-3387/15/5/163},
doi = {10.3390/admsci15050163},
issn = {2076-3387},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Administrative Sciences},
volume = {15},
number = {5},
abstract = {Blockchain technology has transformed modern digital ecosystems by enabling secure, transparent, and automated transactions through smart contracts. However, the increasing complexity of these contracts introduces significant challenges, including high computational costs, scalability limitations, and difficulties in detecting anomalous behavior. In this study, we propose an AI-based optimization framework that enhances the efficiency and security of blockchain smart contracts. The framework integrates Neural Architecture Search (NAS) to automatically design optimal Convolutional Neural Network (CNN) architectures tailored to blockchain data, enabling effective anomaly detection. To address the challenge of limited labeled data, transfer learning is employed to adapt pre-trained CNN models to smart contract patterns, improving model generalization and reducing training time. Furthermore, Model Compression techniques, including filter pruning and quantization, are applied to minimize the computational load, making the framework suitable for deployment in resource-constrained blockchain environments. Experimental results on Ethereum transaction datasets demonstrate that the proposed method achieves significant improvements in anomaly detection accuracy and computational efficiency compared to conventional approaches, offering a practical and scalable solution for smart contract monitoring and optimization.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Blockchain technology has transformed modern digital ecosystems by enabling secure, transparent, and automated transactions through smart contracts. However, the increasing complexity of these contracts introduces significant challenges, including high computational costs, scalability limitations, and difficulties in detecting anomalous behavior. In this study, we propose an AI-based optimization framework that enhances the efficiency and security of blockchain smart contracts. The framework integrates Neural Architecture Search (NAS) to automatically design optimal Convolutional Neural Network (CNN) architectures tailored to blockchain data, enabling effective anomaly detection. To address the challenge of limited labeled data, transfer learning is employed to adapt pre-trained CNN models to smart contract patterns, improving model generalization and reducing training time. Furthermore, Model Compression techniques, including filter pruning and quantization, are applied to minimize the computational load, making the framework suitable for deployment in resource-constrained blockchain environments. Experimental results on Ethereum transaction datasets demonstrate that the proposed method achieves significant improvements in anomaly detection accuracy and computational efficiency compared to conventional approaches, offering a practical and scalable solution for smart contract monitoring and optimization.
4177.
Wang, Xin; Li, Haoyang; Zhang, Zeyang; Chen, Haibo; Zhu, Wenwu
Modular Machine Learning: An Indispensable Path towards New-Generation Large Language Models Technical Report
2025.
@techreport{wang2025modularmachinelearningindispensable,
title = {Modular Machine Learning: An Indispensable Path towards New-Generation Large Language Models},
author = {Xin Wang and Haoyang Li and Zeyang Zhang and Haibo Chen and Wenwu Zhu},
url = {https://arxiv.org/abs/2504.20020},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4176.
Xue, Chao; Li, Jiaxing; Wang, Xiaoxing; Zhan, Yibing; Yan, Junchi; Li, Chun-Guang
On neural architecture search and hyperparameter optimization: A max-flow based approach Journal Article
In: Neural Networks, vol. 188, pp. 107507, 2025, ISSN: 0893-6080.
@article{XUE2025107507,
title = {On neural architecture search and hyperparameter optimization: A max-flow based approach},
author = {Chao Xue and Jiaxing Li and Xiaoxing Wang and Yibing Zhan and Junchi Yan and Chun-Guang Li},
url = {https://www.sciencedirect.com/science/article/pii/S0893608025003867},
doi = {https://doi.org/10.1016/j.neunet.2025.107507},
issn = {0893-6080},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Neural Networks},
volume = {188},
pages = {107507},
abstract = {Automated Machine Learning (AutoML) involves the automatic production of models for specific tasks on given datasets, which can be divided into two aspects: Neural Architecture Search (NAS) for model construction and Hyperparameter Optimization (HPO) for model training. One of the most important components in an AutoML strategy is the search algorithm, which aims to recommend effective configurations according to historical observations. In this work, we propose a novel max-flow based search algorithm for AutoML by representing NAS and HPO as a Max-Flow problem on a graph and thus derive a couple of novel AutoML strategies, dubbed MF-NAS and MF-HPO, which handle the search space and the search strategy graphically. To be specific, MF-NAS induces parallel edges with capacities by combining different operations such as skip connections, convolutions, and pooling, whereas MF-HPO allows parallel edges to be regarded as intervals within the combined search spaces. The learned weights and capacities of the parallel edges are alternately updated during the search process. To make MF-NAS and MF-HPO more efficient, we implement a semi-synchronous search mode for NAS and a warmup scheme for HPO, respectively. We conduct extensive experiments to evaluate the competitive efficacy and efficiency of our proposed MF-NAS and MF-HPO across different datasets and search spaces.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Automated Machine Learning (AutoML) involves the automatic production of models for specific tasks on given datasets, which can be divided into two aspects: Neural Architecture Search (NAS) for model construction and Hyperparameter Optimization (HPO) for model training. One of the most important components in an AutoML strategy is the search algorithm, which aims to recommend effective configurations according to historical observations. In this work, we propose a novel max-flow based search algorithm for AutoML by representing NAS and HPO as a Max-Flow problem on a graph and thus derive a couple of novel AutoML strategies, dubbed MF-NAS and MF-HPO, which handle the search space and the search strategy graphically. To be specific, MF-NAS induces parallel edges with capacities by combining different operations such as skip connections, convolutions, and pooling, whereas MF-HPO allows parallel edges to be regarded as intervals within the combined search spaces. The learned weights and capacities of the parallel edges are alternately updated during the search process. To make MF-NAS and MF-HPO more efficient, we implement a semi-synchronous search mode for NAS and a warmup scheme for HPO, respectively. We conduct extensive experiments to evaluate the competitive efficacy and efficiency of our proposed MF-NAS and MF-HPO across different datasets and search spaces.
4175.
S., Dhivya; S., Prakash; Batumalay, Malathy
Power Quality Assessment in Grid-Connected Solar PV Systems Using Deep Learning Techniques Journal Article
In: Journal of Applied Data Sciences, vol. 6, no. 2, pp. 1192–1208, 2025.
@article{JADS655,
title = {Power Quality Assessment in Grid-Connected Solar PV Systems Using Deep Learning Techniques},
author = {Dhivya S. and Prakash S. and Malathy Batumalay},
url = {https://bright-journal.org/Journal/index.php/JADS/article/view/655},
doi = {10.47738/jads.v6i2.655},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Journal of Applied Data Sciences},
volume = {6},
number = {2},
pages = {1192–1208},
abstract = {To address challenges in stability, power quality, and computational demands while supporting sustainable energy goals in grid-connected solar PV systems, this research introduces a novel deep learning approach: Adaptive Graph-Aware Reinforced Autoencoder with Attention-Based Neural Architecture Search (AGRAAN). AGRAAN simplifies and accelerates the development of neural networks by automatically identifying optimal architectures through Neural Architecture Search (NAS), enabling efficient learning from limited data using Few-Shot Learning, and enhancing performance through attention mechanisms for time-series forecasting. This integrated approach reduces manual tuning and adapts effectively to various tasks. High levels of solar PV integration in power grids introduce variability due to weather conditions and limited forecasting, often resulting in high operational costs. To address this, the AGRAAN model enhances real-time solar variability prediction, improving adaptability, cost-efficiency, and grid stability. NAS supports architectural optimization, Few-Shot Learning improves adaptability with minimal data, and attention mechanisms enhance forecasting accuracy. Additionally, high PV penetration causes voltage fluctuations and harmonic distortions in diverse grid environments. To mitigate these effects, a complementary system named Graph-Aware Reinforced Autoencoder Control System (GRAACS) is proposed. GRAACS detects and manages power quality issues using Autoencoders for anomaly detection, Graph Convolutional Networks (GCNs) for spatial prediction, and Reinforcement Learning for adaptive real-time control. The combined AGRAAN and GRAACS models significantly enhance performance, achieving a high efficiency score of 0.98, an F1-Score of 0.97, and a low Mean Absolute Error (MAE) of 0.11. These results demonstrate the effectiveness of the proposed AI-driven framework in optimizing solar PV grid integration for energy efficiency.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
To address challenges in stability, power quality, and computational demands while supporting sustainable energy goals in grid-connected solar PV systems, this research introduces a novel deep learning approach: Adaptive Graph-Aware Reinforced Autoencoder with Attention-Based Neural Architecture Search (AGRAAN). AGRAAN simplifies and accelerates the development of neural networks by automatically identifying optimal architectures through Neural Architecture Search (NAS), enabling efficient learning from limited data using Few-Shot Learning, and enhancing performance through attention mechanisms for time-series forecasting. This integrated approach reduces manual tuning and adapts effectively to various tasks. High levels of solar PV integration in power grids introduce variability due to weather conditions and limited forecasting, often resulting in high operational costs. To address this, the AGRAAN model enhances real-time solar variability prediction, improving adaptability, cost-efficiency, and grid stability. NAS supports architectural optimization, Few-Shot Learning improves adaptability with minimal data, and attention mechanisms enhance forecasting accuracy. Additionally, high PV penetration causes voltage fluctuations and harmonic distortions in diverse grid environments. To mitigate these effects, a complementary system named Graph-Aware Reinforced Autoencoder Control System (GRAACS) is proposed. GRAACS detects and manages power quality issues using Autoencoders for anomaly detection, Graph Convolutional Networks (GCNs) for spatial prediction, and Reinforcement Learning for adaptive real-time control. The combined AGRAAN and GRAACS models significantly enhance performance, achieving a high efficiency score of 0.98, an F1-Score of 0.97, and a low Mean Absolute Error (MAE) of 0.11. These results demonstrate the effectiveness of the proposed AI-driven framework in optimizing solar PV grid integration for energy efficiency.
4174.
Li, Zhuo; Han, Chengxu
AdaptPerf: On Adaptive and Scalable Computing Power Measurement for Heterogeneous Devices Journal Article
In: IEEE Internet of Things Journal, pp. 1-1, 2025.
@article{10977834,
title = {AdaptPerf: On Adaptive and Scalable Computing Power Measurement for Heterogeneous Devices},
author = {Zhuo Li and Chengxu Han},
url = {https://ieeexplore.ieee.org/abstract/document/10977834},
doi = {10.1109/JIOT.2025.3564899},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Internet of Things Journal},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4173.
Hou, Boyu; Feng, Liang; Chen, Xuefeng; Tang, Jing; Tan, Kay Chen; Liao, Xiaofeng
Evolutionary Transfer Neural Architecture Search Across Spaces via Representation Learning Journal Article
In: IEEE Transactions on Evolutionary Computation, pp. 1-1, 2025.
@article{10979972,
title = {Evolutionary Transfer Neural Architecture Search Across Spaces via Representation Learning},
author = {Boyu Hou and Liang Feng and Xuefeng Chen and Jing Tang and Kay Chen Tan and Xiaofeng Liao},
url = {https://ieeexplore.ieee.org/abstract/document/10979972},
doi = {10.1109/TEVC.2025.3565326},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Evolutionary Computation},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4172.
Li, Yangyang; Liu, Guanlong; Shang, Ronghua; Jiao, Licheng
Meta knowledge assisted Evolutionary Neural Architecture Search Technical Report
2025.
@techreport{li2025metaknowledgeassistedevolutionary,
title = {Meta knowledge assisted Evolutionary Neural Architecture Search},
author = {Yangyang Li and Guanlong Liu and Ronghua Shang and Licheng Jiao},
url = {https://arxiv.org/abs/2504.21545},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4171.
Chen, Zeyang; Xu, Dezhi; Shen, Chao; Ye, Yujian; Jiang, Bin
Miniature Real-Time Compact Deep Neural Network With Zero-Shot Neural Architecture Search for Lithium-Ion Battery Fault Diagnosis Journal Article
In: IEEE Transactions on Industrial Informatics, pp. 1-10, 2025.
@article{10980014,
title = {Miniature Real-Time Compact Deep Neural Network With Zero-Shot Neural Architecture Search for Lithium-Ion Battery Fault Diagnosis},
author = {Zeyang Chen and Dezhi Xu and Chao Shen and Yujian Ye and Bin Jiang},
url = {https://ieeexplore.ieee.org/abstract/document/10980014},
doi = {10.1109/TII.2025.3556081},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Transactions on Industrial Informatics},
pages = {1-10},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
4170.
Wang, Sixuan; Yin, Jiao; Cao, Jinli; Tang, MingJian; Wang, Hua; Zhang, Yanchun
ABG-NAS: Adaptive Bayesian Genetic Neural Architecture Search for Graph Representation Learning Technical Report
2025.
@techreport{wang2025abgnasadaptivebayesiangenetic,
title = {ABG-NAS: Adaptive Bayesian Genetic Neural Architecture Search for Graph Representation Learning},
author = {Sixuan Wang and Jiao Yin and Jinli Cao and MingJian Tang and Hua Wang and Yanchun Zhang},
url = {https://arxiv.org/abs/2504.21254},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4169.
Moon, JunGi; Jung, SangJin; Suh, SungMin; Pyo, JongCheol
Development of deep learning quantization framework for remote sensing edge device to estimate inland water quality in South Korea Journal Article
In: Water Research, vol. 283, pp. 123760, 2025, ISSN: 0043-1354.
@article{MOON2025123760,
title = {Development of deep learning quantization framework for remote sensing edge device to estimate inland water quality in South Korea},
author = {JunGi Moon and SangJin Jung and SungMin Suh and JongCheol Pyo},
url = {https://www.sciencedirect.com/science/article/pii/S0043135425006694},
doi = {https://doi.org/10.1016/j.watres.2025.123760},
issn = {0043-1354},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Water Research},
volume = {283},
pages = {123760},
abstract = {Recent achievements in the fields of deep learning and remote sensing have led to their application in monitoring river water quality. One of the most researched methods is the estimation of total suspended solid (TSS) concentrations using multispectral imagery and convolutional neural network (CNN) models. Owing to the sorption capacity of other pollutants, TSS monitoring is essential. However, despite recent advances in deep learning, the application of contemporary technologies in water quality monitoring has not yet been fully explored. This study aims to develop a framework for on-device AI that can be applied to edge devices through quantization using a lightweight deep learning model. Lightweight CNN models were identified using neural architecture search (NAS) in conjunction with Pareto optimization, achieving high performance (0.806 of Nash-Sutcliffe efficiency (NSE)) while minimizing computational burden (8.118 MB). The model sizes were further compressed (0.736 MB) through the application of post-training quantization (PTQ) and quantization aware training (QAT), ensuring that accuracy (0.831 of NSE) was preserved. This provides a scalable approach for real-time TSS monitoring, bridging the gap between advanced deep learning techniques and practical environmental applications. These applications indicate that it is possible to estimate other water quality indices using multispectral imagery. It enables the tracing of the source of contamination and facilitates rapid responses by identifying changes in real time.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Recent achievements in the fields of deep learning and remote sensing have led to their application in monitoring river water quality. One of the most researched methods is the estimation of total suspended solid (TSS) concentrations using multispectral imagery and convolutional neural network (CNN) models. Owing to the sorption capacity of other pollutants, TSS monitoring is essential. However, despite recent advances in deep learning, the application of contemporary technologies in water quality monitoring has not yet been fully explored. This study aims to develop a framework for on-device AI that can be applied to edge devices through quantization using a lightweight deep learning model. Lightweight CNN models were identified using neural architecture search (NAS) in conjunction with Pareto optimization, achieving high performance (0.806 of Nash-Sutcliffe efficiency (NSE)) while minimizing computational burden (8.118 MB). The model sizes were further compressed (0.736 MB) through the application of post-training quantization (PTQ) and quantization aware training (QAT), ensuring that accuracy (0.831 of NSE) was preserved. This provides a scalable approach for real-time TSS monitoring, bridging the gap between advanced deep learning techniques and practical environmental applications. These applications indicate that it is possible to estimate other water quality indices using multispectral imagery. It enables the tracing of the source of contamination and facilitates rapid responses by identifying changes in real time.
4168.
Li, Nan; Ma, Lianbo; Wang, Rui; Cheng, Shi; Sun, Yanan; Xue, Bing; Zhang, Mengjie
Listwise ranking predictor for evolutionary neural architecture search Journal Article
In: Swarm and Evolutionary Computation, vol. 96, pp. 101956, 2025, ISSN: 2210-6502.
@article{LI2025101956,
title = {Listwise ranking predictor for evolutionary neural architecture search},
author = {Nan Li and Lianbo Ma and Rui Wang and Shi Cheng and Yanan Sun and Bing Xue and Mengjie Zhang},
url = {https://www.sciencedirect.com/science/article/pii/S2210650225001142},
doi = {https://doi.org/10.1016/j.swevo.2025.101956},
issn = {2210-6502},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Swarm and Evolutionary Computation},
volume = {96},
pages = {101956},
abstract = {In evolutionary neural architecture search (ENAS), the accuracy predictors (i.e., regression models) have been successfully applied to save computational costs for the evaluation of network architectures. However, the accuracy of these predictors is largely limited by the small amount of evaluated architectures that may be difficult to obtain. Such accuracy predictors with prediction bias often lead to an inaccurate ranking, misleading the selection of ENAS. To alleviate the above limitations, we design an efficient and novel listwise ranking predictor (LRP) for ENAS to directly predict the ranking of each architecture instead of the numerical accuracy value of each architecture. Specifically, the training data is constructed by the proposed random encoding-combination (REC) strategy, which can generate substantial training data using the small number of evaluated architectures (data level). These specially constructed training data are used to train LRP, which can convert the complex regression task into a ranking task to reduce ranking bias (model level). The proposed NAS method is compared with state-of-the-art NAS methods on widely-used benchmark datasets and practical application. Experimental results demonstrate that LRP can alleviate the ranking disorder problem and outperform others in terms of both effectiveness and efficiency.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In evolutionary neural architecture search (ENAS), the accuracy predictors (i.e., regression models) have been successfully applied to save computational costs for the evaluation of network architectures. However, the accuracy of these predictors is largely limited by the small amount of evaluated architectures that may be difficult to obtain. Such accuracy predictors with prediction bias often lead to an inaccurate ranking, misleading the selection of ENAS. To alleviate the above limitations, we design an efficient and novel listwise ranking predictor (LRP) for ENAS to directly predict the ranking of each architecture instead of the numerical accuracy value of each architecture. Specifically, the training data is constructed by the proposed random encoding-combination (REC) strategy, which can generate substantial training data using the small number of evaluated architectures (data level). These specially constructed training data are used to train LRP, which can convert the complex regression task into a ranking task to reduce ranking bias (model level). The proposed NAS method is compared with state-of-the-art NAS methods on widely-used benchmark datasets and practical application. Experimental results demonstrate that LRP can alleviate the ranking disorder problem and outperform others in terms of both effectiveness and efficiency.
4167.
Li, Nan; Ma, Lianbo; Wang, Rui; Cheng, Shi; Sun, Yanan; Xue, Bing; Zhang, Mengjie
Listwise ranking predictor for evolutionary neural architecture search Journal Article
In: Swarm and Evolutionary Computation, vol. 96, pp. 101956, 2025, ISSN: 2210-6502.
@article{LI2025101956b,
title = {Listwise ranking predictor for evolutionary neural architecture search},
author = {Nan Li and Lianbo Ma and Rui Wang and Shi Cheng and Yanan Sun and Bing Xue and Mengjie Zhang},
url = {https://www.sciencedirect.com/science/article/pii/S2210650225001142},
doi = {https://doi.org/10.1016/j.swevo.2025.101956},
issn = {2210-6502},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {Swarm and Evolutionary Computation},
volume = {96},
pages = {101956},
abstract = {In evolutionary neural architecture search (ENAS), the accuracy predictors (i.e., regression models) have been successfully applied to save computational costs for the evaluation of network architectures. However, the accuracy of these predictors is largely limited by the small amount of evaluated architectures that may be difficult to obtain. Such accuracy predictors with prediction bias often lead to an inaccurate ranking, misleading the selection of ENAS. To alleviate the above limitations, we design an efficient and novel listwise ranking predictor (LRP) for ENAS to directly predict the ranking of each architecture instead of the numerical accuracy value of each architecture. Specifically, the training data is constructed by the proposed random encoding-combination (REC) strategy, which can generate substantial training data using the small number of evaluated architectures (data level). These specially constructed training data are used to train LRP, which can convert the complex regression task into a ranking task to reduce ranking bias (model level). The proposed NAS method is compared with state-of-the-art NAS methods on widely-used benchmark datasets and practical application. Experimental results demonstrate that LRP can alleviate the ranking disorder problem and outperform others in terms of both effectiveness and efficiency.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In evolutionary neural architecture search (ENAS), the accuracy predictors (i.e., regression models) have been successfully applied to save computational costs for the evaluation of network architectures. However, the accuracy of these predictors is largely limited by the small amount of evaluated architectures that may be difficult to obtain. Such accuracy predictors with prediction bias often lead to an inaccurate ranking, misleading the selection of ENAS. To alleviate the above limitations, we design an efficient and novel listwise ranking predictor (LRP) for ENAS to directly predict the ranking of each architecture instead of the numerical accuracy value of each architecture. Specifically, the training data is constructed by the proposed random encoding-combination (REC) strategy, which can generate substantial training data using the small number of evaluated architectures (data level). These specially constructed training data are used to train LRP, which can convert the complex regression task into a ranking task to reduce ranking bias (model level). The proposed NAS method is compared with state-of-the-art NAS methods on widely-used benchmark datasets and practical application. Experimental results demonstrate that LRP can alleviate the ranking disorder problem and outperform others in terms of both effectiveness and efficiency.
4166.
Guarrasi, Valerio; Mogensen, Klara; Tassinari, Sara; Qvarlander, Sara; Soda, Paolo
Timing Is Everything: Finding the Optimal Fusion Points in Multimodal Medical Imaging Technical Report
2025.
@techreport{guarrasi2025timingeverythingfindingoptimal,
title = {Timing Is Everything: Finding the Optimal Fusion Points in Multimodal Medical Imaging},
author = {Valerio Guarrasi and Klara Mogensen and Sara Tassinari and Sara Qvarlander and Paolo Soda},
url = {https://arxiv.org/abs/2505.02467},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
4165.
Wani, M. Arif; Sultan, Bisma; Ali, Sarwat; Sofi, Mukhtar Ahmad
Evolutionary Algorithm-Based Neural Architecture Search Book Chapter
In: Advances in Deep Learning, Volume 2, pp. 15–30, Springer Nature Singapore, Singapore, 2025, ISBN: 978-981-96-3498-9.
@inbook{Wani2025,
title = {Evolutionary Algorithm-Based Neural Architecture Search},
author = {M. Arif Wani and Bisma Sultan and Sarwat Ali and Mukhtar Ahmad Sofi},
url = {https://doi.org/10.1007/978-981-96-3498-9_2},
doi = {10.1007/978-981-96-3498-9_2},
isbn = {978-981-96-3498-9},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
booktitle = {Advances in Deep Learning, Volume 2},
pages = {15–30},
publisher = {Springer Nature Singapore},
address = {Singapore},
abstract = {The increasing demand for specialized neural network architectures that cater to specific tasks has given rise to automated methods for architecture design, alleviating the need for manual, labor-intensive processes. Neural Architecture Search (NAS) has emerged as a key solution, enabling the discovery of optimized neural networks without human intervention. Among the various approaches to NAS, Evolutionary Algorithm-based NAS has proven to be particularly effective due to its ability to efficiently navigate the vast search space of neural architectures by employing biologically inspired optimization techniques.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The increasing demand for specialized neural network architectures that cater to specific tasks has given rise to automated methods for architecture design, alleviating the need for manual, labor-intensive processes. Neural Architecture Search (NAS) has emerged as a key solution, enabling the discovery of optimized neural networks without human intervention. Among the various approaches to NAS, Evolutionary Algorithm-based NAS has proven to be particularly effective due to its ability to efficiently navigate the vast search space of neural architectures by employing biologically inspired optimization techniques.