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.
2023
Deng, Yuwen; Kang, Wang; Xing, Wei W.
Differentiable Multi-Fidelity Fusion: Efficient Learning of Physics Simulations with Neural Architecture Search and Transfer Learning Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-06904,
title = {Differentiable Multi-Fidelity Fusion: Efficient Learning of Physics Simulations with Neural Architecture Search and Transfer Learning},
author = {Yuwen Deng and Wang Kang and Wei W. Xing},
url = {https://doi.org/10.48550/arXiv.2306.06904},
doi = {10.48550/arXiv.2306.06904},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.06904},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Jidney, Tasmia Tahmida; Biswas, Angona; Nasim, Md Abdullah Al; Hossain, Ismail; Alam, Md Jahangir; Talukder, Sajedul; Hossain, Mofazzal; Ullah, Md. Azim
AutoML Systems For Medical Imaging Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-04750,
title = {AutoML Systems For Medical Imaging},
author = {Tasmia Tahmida Jidney and Angona Biswas and Md Abdullah Al Nasim and Ismail Hossain and Md Jahangir Alam and Sajedul Talukder and Mofazzal Hossain and Md. Azim Ullah},
url = {https://doi.org/10.48550/arXiv.2306.04750},
doi = {10.48550/arXiv.2306.04750},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.04750},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Salinas-Guerra, Rocío; Mejía-Dios, Jesús-Adolfo; Mezura-Montes, Efrén; Márquez-Grajales, Aldo
An Evolutionary Bilevel Optimization Approach for Neuroevolution Book Chapter
In: Castillo, Oscar; Melin, Patricia (Ed.): Hybrid Intelligent Systems Based on Extensions of Fuzzy Logic, Neural Networks and Metaheuristics, pp. 395–423, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-28999-6.
@inbook{Salinas-Guerra2023,
title = {An Evolutionary Bilevel Optimization Approach for Neuroevolution},
author = {Rocío Salinas-Guerra and Jesús-Adolfo Mejía-Dios and Efrén Mezura-Montes and Aldo Márquez-Grajales},
editor = {Oscar Castillo and Patricia Melin},
url = {https://doi.org/10.1007/978-3-031-28999-6_25},
doi = {10.1007/978-3-031-28999-6_25},
isbn = {978-3-031-28999-6},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Hybrid Intelligent Systems Based on Extensions of Fuzzy Logic, Neural Networks and Metaheuristics},
pages = {395–423},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Convolutional neural networks (CNN) have been extensively studied and achieved significant progress on a variety of computer vision tasks in recent years. However, the design of their architectures remains challenging due to the computational cost and the number of parameters used. Neuroevolution has offered various evolutionary algorithms to provide a suitable option for designing CNNs. Moreover, modeling such a design as a bilevel optimization problem has recently attracted the interest of researchers and practitioners because it can be seen as a hierarchical task. This work precisely addresses it as a bilevel optimization problem. Unlike existing approaches, the upper level minimizes the complexity of the network (described by the number of its parameters), while the lower level optimizes the topology of the network structure for maximum accuracy. The results suggest that based on user preferences, the bilevel optimization approach can report neural architecture with higher accuracy values or simpler convolutional neural networks.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Chauhan, Vinod Kumar; Zhou, Jiandong; Lu, Ping; Molaei, Soheila; Clifton, David A.
A Brief Review of Hypernetworks in Deep Learning Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-06955,
title = {A Brief Review of Hypernetworks in Deep Learning},
author = {Vinod Kumar Chauhan and Jiandong Zhou and Ping Lu and Soheila Molaei and David A. Clifton},
url = {https://doi.org/10.48550/arXiv.2306.06955},
doi = {10.48550/arXiv.2306.06955},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.06955},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Richey, Blake; Clay, Mitchell; Grecos, Christos; Shirvaikar, Mukul
Evolution of hardware-aware neural architecture search (NAS) on the edge Proceedings Article
In: Kehtarnavaz, Nasser; Shirvaikar, Mukul V. (Ed.): Real-Time Image Processing and Deep Learning 2023, pp. 125280A, International Society for Optics and Photonics SPIE, 2023.
@inproceedings{10.1117/12.2664894,
title = {Evolution of hardware-aware neural architecture search (NAS) on the edge},
author = {Blake Richey and Mitchell Clay and Christos Grecos and Mukul Shirvaikar},
editor = {Nasser Kehtarnavaz and Mukul V. Shirvaikar},
url = {https://doi.org/10.1117/12.2664894},
doi = {10.1117/12.2664894},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Real-Time Image Processing and Deep Learning 2023},
volume = {12528},
pages = {125280A},
publisher = {SPIE},
organization = {International Society for Optics and Photonics},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Lei, Tongfei; Hu, Jiabei; Riaz, Saleem
An innovative approach based on meta-learning for real-time modal fault diagnosis with small sample learning Journal Article
In: Frontiers in Physics, vol. 11, 2023, ISSN: 2296-424X.
@article{10.3389/fphy.2023.1207381,
title = {An innovative approach based on meta-learning for real-time modal fault diagnosis with small sample learning},
author = {Tongfei Lei and Jiabei Hu and Saleem Riaz},
url = {https://www.frontiersin.org/articles/10.3389/fphy.2023.1207381},
doi = {10.3389/fphy.2023.1207381},
issn = {2296-424X},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Frontiers in Physics},
volume = {11},
abstract = {The actual multimodal process data usually exhibit non-linear time correlation and non-Gaussian distribution accompanied by new modes. Existing fault diagnosis methods have difficulty adapting to the complex nature of new modalities and are unable to train models based on small samples. Therefore, this paper proposes a new modal fault diagnosis method based on meta-learning (ML) and neural architecture search (NAS), MetaNAS. Specifically, the best performing network model of the existing modal is first automatically obtained using NAS, and then, the fault diagnosis model design is learned from the NAS of the existing model using ML. Finally, when generating new modalities, the gradient is updated based on the learned design experience, i.e., new modal fault diagnosis models are quickly generated under small sample conditions. The effectiveness and feasibility of the proposed method are fully verified by the numerical system and simulation experiments of the Tennessee Eastman (TE) chemical process. As a primary goal, the abstract should render the general significance and conceptual advance of the work clearly accessible to a broad readership. References should not be cited in the abstract. Leave the Abstract empty if your article does not require one–please see the “Article types” on every Frontiers journal page for full details.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Kang, Beom Woo; Wohn, Junho; Lee, Seongju; Park, Sunghyun; Noh, Yung-Kyun; Park, Yongjun
Synchronization-Aware NAS for an Efficient Collaborative Inference on Mobile Platforms Proceedings Article
In: Proceedings of the 24th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems, pp. 13–25, Association for Computing Machinery, Orlando, FL, USA, 2023, ISBN: 9798400701740.
@inproceedings{10.1145/3589610.3596284,
title = {Synchronization-Aware NAS for an Efficient Collaborative Inference on Mobile Platforms},
author = {Beom Woo Kang and Junho Wohn and Seongju Lee and Sunghyun Park and Yung-Kyun Noh and Yongjun Park},
url = {https://doi.org/10.1145/3589610.3596284},
doi = {10.1145/3589610.3596284},
isbn = {9798400701740},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Proceedings of the 24th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems},
pages = {13–25},
publisher = {Association for Computing Machinery},
address = {Orlando, FL, USA},
series = {LCTES 2023},
abstract = {Previous neural architecture search (NAS) approaches for mobile platforms have achieved great success in designing a slim-but-accurate neural network that is generally well-matched to a single computing unit such as a CPU or GPU. However, as recent mobile devices consist of multiple heterogeneous computing units, the next main challenge is to maximize both accuracy and efficiency by fully utilizing multiple available resources. We propose an ensemble-like approach with intermediate feature aggregations, namely synchronizations, for active collaboration between individual models on a mobile device. A main challenge is to determine the optimal synchronization strategies for achieving both performance and efficiency. To this end, we propose SyncNAS to automate the exploration of synchronization strategies for collaborative neural architectures that maximize utilization of heterogeneous computing units on a target device. We introduce a novel search space for synchronization strategy and apply Monte Carlo tree search (MCTS) algorithm to improve the sampling efficiency and reduce the search cost. On ImageNet, our collaborative model based on MobileNetV2 achieves 2.7% top-1 accuracy improvement within the baseline latency budget. Under the reduced target latency down to half, our model maintains higher accuracy than its baseline model, owing to the enhanced utilization and collaboration. As an impact of MCTS, SyncNAS reduces its search cost by up to 21x in searching for the optimal strategy.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Qin, Shixi; Zhang, Zixun; Jiang, Yuncheng; Cui, Shuguang; Cheng, Shenghui; Li, Zhen
NG-NAS: Node growth neural architecture search for 3D medical image segmentation Journal Article
In: Computerized Medical Imaging and Graphics, vol. 108, pp. 102268, 2023, ISSN: 0895-6111.
@article{QIN2023102268,
title = {NG-NAS: Node growth neural architecture search for 3D medical image segmentation},
author = {Shixi Qin and Zixun Zhang and Yuncheng Jiang and Shuguang Cui and Shenghui Cheng and Zhen Li},
url = {https://www.sciencedirect.com/science/article/pii/S0895611123000861},
doi = {https://doi.org/10.1016/j.compmedimag.2023.102268},
issn = {0895-6111},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Computerized Medical Imaging and Graphics},
volume = {108},
pages = {102268},
abstract = {Neural architecture search (NAS) has been applied to design proper 3D networks for medical image segmentation. In order to reduce the computation cost in NAS, researchers tend to adopt weight sharing mechanism to search architectures in a supernet. However, recent studies state that the searched architecture rankings may not be accurate with weight sharing mechanism because the training situations are inconsistent between the searching and training phases. In addition, some NAS algorithms design inflexible supernets that only search operators in a pre-defined backbone and ignore the importance of network topology, which limits the performance of searched architecture. To avoid weight sharing mechanism which may lead to inaccurate results and to comprehensively search network topology and operators, we propose a novel NAS algorithm called NG-NAS. Following the previous studies, we consider the segmentation network as a U-shape structure composed of a set of nodes. Instead of searching from the supernet with a limited search space, our NG-NAS starts from a simple architecture with only 5 nodes, and greedily grows the best candidate node until meeting the constraint. We design 2 kinds of node generations to form various network topological structures and prepare 4 candidate operators for each node. To efficiently evaluate candidate node generations, we use NAS without training strategies. We evaluate our method on several public 3D medical image segmentation benchmarks and achieve state-of-the-art performance, demonstrating the effectiveness of the searched architecture and our NG-NAS. Concretely, our method achieves an average Dice score of 85.11 on MSD liver, 65.70 on MSD brain, and 87.59 in BTCV, which performs much better than the previous SOTA methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Jin, Xiao; Mu, Xin-Yue; Xu, Jing
Searching for the Fakes: Efficient Neural Architecture Search for General Face Forgery Detection Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-08830,
title = {Searching for the Fakes: Efficient Neural Architecture Search for General Face Forgery Detection},
author = {Xiao Jin and Xin-Yue Mu and Jing Xu},
url = {https://doi.org/10.48550/arXiv.2306.08830},
doi = {10.48550/arXiv.2306.08830},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.08830},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Yüzügüler, Ahmet Caner; Dimitriadis, Nikolaos; Frossard, Pascal
Flexible Channel Dimensions for Differentiable Architecture Search Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-08021,
title = {Flexible Channel Dimensions for Differentiable Architecture Search},
author = {Ahmet Caner Yüzügüler and Nikolaos Dimitriadis and Pascal Frossard},
url = {https://doi.org/10.48550/arXiv.2306.08021},
doi = {10.48550/arXiv.2306.08021},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.08021},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Ji, Junzhong; Wang, Xingyu
Fast Progressive Differentiable Architecture Search based on adaptive task granularity reorganization Journal Article
In: Information Sciences, vol. 645, pp. 119326, 2023, ISSN: 0020-0255.
@article{JI2023119326,
title = {Fast Progressive Differentiable Architecture Search based on adaptive task granularity reorganization},
author = {Junzhong Ji and Xingyu Wang},
url = {https://www.sciencedirect.com/science/article/pii/S0020025523009118},
doi = {https://doi.org/10.1016/j.ins.2023.119326},
issn = {0020-0255},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Information Sciences},
volume = {645},
pages = {119326},
abstract = {Shrinkage methods reduce the search space of a Differentiable Architecture Search (DARTS) by progressively discarding candidates, which accelerates the search speed. However, their shrinkage strategy suffers from the vulnerability of too fine task granularity. In other words, they drop only one of the least promising candidates per round of shrinkage, which is suboptimal in terms of performance and efficiency. In this study, we introduce the concept of Granular Computing (GrC) into the shrinkage method and present a Fast Progressive Differentiable Architecture Search (FP-DARTS) method. This method effectively reduces the computational complexity of each round of shrinkage, thereby improving the efficiency and performance of the algorithm. FP-DARTS can be divided into three stages: adaptive granularity division and selection, granular-channel performance evaluation, and progressive shrinkage. In the first stage, to reorganize the task granularity, we cluster the candidate operations into granular-channels and adaptively select the appropriate task granularity. We also propose a dynamic clustering strategy to avoid introducing additional computation. In the second stage, we train the architecture parameters to measure the potential of the granular-channels. In the third stage, to improve the stability of the shrinkage results, we introduce a channel annealing mechanism to smoothly discard unpromising granular-channels. We conducted systematic experiments on CIFAR-10 and ImageNet and achieved a test accuracy of 97.56% on CIFAR-10 with 0.04 GPU-days, and a test accuracy of 75.5% on ImageNet with 1.2 GPU-days. We also conducted experiments on the search space of NAS-Bench-201, and obtained test accuracies of 94.22, 73.07, and 46.23% for CIFAR-10, CIFAR-100 and ImageNet16-120, respectively. The above experimental results demonstrate that FP-DARTS achieves higher search speed and competitive performance compared to other state-of-the-art shrinkage methods and non-shrinkage methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Eustratiadis, Panagiotis; Dudziak, Lukasz; Li, Da; Hospedales, Timothy M.
Neural Fine-Tuning Search for Few-Shot Learning Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-09295,
title = {Neural Fine-Tuning Search for Few-Shot Learning},
author = {Panagiotis Eustratiadis and Lukasz Dudziak and Da Li and Timothy M. Hospedales},
url = {https://doi.org/10.48550/arXiv.2306.09295},
doi = {10.48550/arXiv.2306.09295},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.09295},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Guo, Bicheng; Xu, Lilin; Chen, Tao; Ye, Peng; He, Shibo; Liu, Haoyu; Chen, Jiming
Latency-aware Neural Architecture Performance Predictor with Query-to-Tier Technique Journal Article
In: IEEE Transactions on Circuits and Systems for Video Technology, pp. 1-1, 2023.
@article{10155437,
title = {Latency-aware Neural Architecture Performance Predictor with Query-to-Tier Technique},
author = {Bicheng Guo and Lilin Xu and Tao Chen and Peng Ye and Shibo He and Haoyu Liu and Jiming Chen},
doi = {10.1109/TCSVT.2023.3287684},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Jeon, Seunghyeok; Choi, Yonghun; Cho, Yeonwoo; Cha, Hojung
HarvNet: Resource-Optimized Operation of Multi-Exit Deep Neural Networks on Energy Harvesting Devices Proceedings Article
In: Proceedings of the 21st Annual International Conference on Mobile Systems, Applications and Services, pp. 42–55, Association for Computing Machinery, Helsinki, Finland, 2023, ISBN: 9798400701108.
@inproceedings{10.1145/3581791.3596845,
title = {HarvNet: Resource-Optimized Operation of Multi-Exit Deep Neural Networks on Energy Harvesting Devices},
author = {Seunghyeok Jeon and Yonghun Choi and Yeonwoo Cho and Hojung Cha},
url = {https://doi.org/10.1145/3581791.3596845},
doi = {10.1145/3581791.3596845},
isbn = {9798400701108},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Proceedings of the 21st Annual International Conference on Mobile Systems, Applications and Services},
pages = {42–55},
publisher = {Association for Computing Machinery},
address = {Helsinki, Finland},
series = {MobiSys '23},
abstract = {Optimizing deep neural networks (DNNs) running on resource-constrained devices, such as energy harvesting sensor devices, poses unique challenges due to the limited memory and varying energy conditions. Existing efforts have shown that deploying a multi-exit network mitigates the problem by allowing tradeoffs between accuracy and computational complexity. However, previous works did not fully consider two essential requirements: optimized neural architecture and optimized inference policy. In this paper, we present HarvNet, which comprises two complementary techniques for generating and operating a multi-exit network for energy harvesting devices. First, we provide a neural architecture search scheme, HarvNAS, which configures the best multi-exit architecture while meeting memory and energy constraints. Second, HarvSched learns and constructs the best progressive inference policy with different energy constraints by considering runtime factors, such as the harvesting status and the energy storage level. We implemented HarvNAS using the TensorFlow framework and then implemented and evaluated HarvSched on an MSP430-based sensor device. The evaluation showed that HarvNAS generated a model with up to 2.6%p higher accuracy while saving up to 70% of memory compared to the existing technique, and HarvSched enabled zero-downtime operation of the generated model.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Park, Seonghoon; Cho, Yeonwoo; Jun, Hyungchol; Lee, Jeho; Cha, Hojung
OmniLive: Super-Resolution Enhanced 360° Video Live Streaming for Mobile Devices Proceedings Article
In: Proceedings of the 21st Annual International Conference on Mobile Systems, Applications and Services, pp. 261–274, Association for Computing Machinery, Helsinki, Finland, 2023, ISBN: 9798400701108.
@inproceedings{10.1145/3581791.3596851,
title = {OmniLive: Super-Resolution Enhanced 360° Video Live Streaming for Mobile Devices},
author = {Seonghoon Park and Yeonwoo Cho and Hyungchol Jun and Jeho Lee and Hojung Cha},
url = {https://doi.org/10.1145/3581791.3596851},
doi = {10.1145/3581791.3596851},
isbn = {9798400701108},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Proceedings of the 21st Annual International Conference on Mobile Systems, Applications and Services},
pages = {261–274},
publisher = {Association for Computing Machinery},
address = {Helsinki, Finland},
series = {MobiSys '23},
abstract = {The live streaming of omnidirectional video (ODV) on mobile devices demands considerable network resources; thus, current mobile networks are incapable of providing users with high-quality ODV equivalent to conventional flat videos. We observe that mobile devices, in fact, underutilize graphics processing units (GPUs) while processing ODVs; hence, we envisage an opportunity exists in exploiting video super-resolution (VSR) for improved ODV quality. However, the device-specific discrepancy in GPU capability and dynamic behavior of GPU frequency in mobile devices create a challenge in providing VSR-enhanced ODV streaming. In this paper, we propose OmniLive, an on-device VSR system for mobile ODV live streaming. OmniLive addresses the dynamicity of GPU capability with an anytime inference-based VSR technique called Omni SR. For Omni SR, we design a VSR deep neural network (DNN) model with multiple exits and an inference scheduler that decides on the exit of the model at runtime. OmniLive also solves the performance heterogeneity of mobile GPUs using the Omni neural architecture search (NAS) scheme. Omni NAS finds an appropriate DNN model for each mobile device with Omni SR-specific neural architecture search techniques. We implemented OmniLive as a fully functioning system encompassing a streaming server and Android application. The experiment results show that our anytime VSR model provides four times upscaled videos while saving up to 57.15% of inference time compared with the previous super-resolution model showing the lowest inference time on mobile devices. Moreover, OmniLive can maintain 30 frames per second while fully utilizing GPUs on various mobile devices.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Mukhopadhyay, Shalini; Dey, Swarnava; Ghose, Avik
TinyPuff: Automated Design of Tiny Smoking Puff Classifiers for Body Worn Devices Proceedings Article
In: Proceedings of the 8th Workshop on Body-Centric Computing Systems, pp. 7–12, Association for Computing Machinery, Helsinki, Finland, 2023, ISBN: 9798400702112.
@inproceedings{10.1145/3597061.3597259,
title = {TinyPuff: Automated Design of Tiny Smoking Puff Classifiers for Body Worn Devices},
author = {Shalini Mukhopadhyay and Swarnava Dey and Avik Ghose},
url = {https://doi.org/10.1145/3597061.3597259},
doi = {10.1145/3597061.3597259},
isbn = {9798400702112},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Proceedings of the 8th Workshop on Body-Centric Computing Systems},
pages = {7–12},
publisher = {Association for Computing Machinery},
address = {Helsinki, Finland},
series = {BodySys '23},
abstract = {Smoking is a significant cause of death and deterioration of health worldwide, affecting active and passive smokers. Cessation of smoking contributes to an essential health and wellness application owing to the broad range of health problems such as cancer, hypertension, and several cardiopulmonary diseases. Personalized smoking-cessation applications can be very effective in helping users to stop smoking if there are detections and interventions done at the right time. This requires real-time detection of smoking puffs. Such applications are made feasible by day-long monitoring and smoking puff detection from unobtrusive devices such as wearables. This paper proposes a deep inference technique for the real-time detection of smoking puffs on a wearable device. We show that a simple, sequential Convolutional Neural Network (CNN) using only 6-axis Inertial signals can be utilized in place of complex and resource-consuming Deep Learning models. The accuracy achieved is comparable to State-of-the-Art techniques with an F1 score of 0.81, although the model size is tiny - 114 kB. Such small models can be deployed on the lowest configuration hardware platforms, achieving accurate but high-speed, low-power inference on conventional smartwatches. We ensure that the auto-designed models are directly compatible with resource-constrained platforms such as TensorFlow Lite and TensorFlow Lite for Microcontrollers (TFLM) without requiring further use of model reduction and optimization techniques. Our proposed approach will allow affordable wearable device manufacturers to run smoking detection on their devices, as it is tiny enough to fit TinyML platforms and is only dependent on IMU sensors that are universally available.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Sun, Haiyang; Zhang, Fulin; Lian, Zheng; Guo, Yingying; Zhang, Shilei
MFAS: Emotion Recognition through Multiple Perspectives Fusion Architecture Search Emulating Human Cognition Technical Report
2023.
@techreport{sun2023mfas,
title = {MFAS: Emotion Recognition through Multiple Perspectives Fusion Architecture Search Emulating Human Cognition},
author = {Haiyang Sun and Fulin Zhang and Zheng Lian and Yingying Guo and Shilei Zhang},
url = {https://arxiv.org/abs/2306.09361},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Javan, Mehraveh; Toews, Matthew; Pedersoli, Marco
Balanced Mixture of SuperNets for Learning the CNN Pooling Architecture Technical Report
2023.
@techreport{javan2023balanced,
title = {Balanced Mixture of SuperNets for Learning the CNN Pooling Architecture},
author = {Mehraveh Javan and Matthew Toews and Marco Pedersoli},
url = {https://arxiv.org/abs/2306.11982},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Schalk, Daan; Sommen, Fons; Sanberg, Willem P.
Radar-based Object Classification in ADAS with Hardware-Aware NAS and Input Region Scaling Proceedings Article
In: 2023 IEEE Radar Conference (RadarConf23), pp. 1-6, 2023.
@inproceedings{10149615,
title = {Radar-based Object Classification in ADAS with Hardware-Aware NAS and Input Region Scaling},
author = {Daan Schalk and Fons Sommen and Willem P. Sanberg},
url = {https://ieeexplore.ieee.org/abstract/document/10149615},
doi = {10.1109/RadarConf2351548.2023.10149615},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE Radar Conference (RadarConf23)},
pages = {1-6},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Kim, Sehoon; Hooper, Coleman; Wattanawong, Thanakul; Kang, Minwoo; Yan, Ruohan; Genc, Hasan; Dinh, Grace; Huang, Qijing; Keutzer, Kurt; Mahoney, Michael W.; Shao, Sophia; Gholami, Amir
Full Stack Optimization of Transformer Inference Proceedings Article
In: Architecture and System Support for Transformer Models (ASSYST @ISCA 2023), 2023.
@inproceedings{<LineBreak>kim2023full,
title = {Full Stack Optimization of Transformer Inference},
author = {Sehoon Kim and Coleman Hooper and Thanakul Wattanawong and Minwoo Kang and Ruohan Yan and Hasan Genc and Grace Dinh and Qijing Huang and Kurt Keutzer and Michael W. Mahoney and Sophia Shao and Amir Gholami},
url = {https://openreview.net/forum?id=GtyQbLUUagE},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Architecture and System Support for Transformer Models (ASSYST @ISCA 2023)},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Liu, Jiaqi; Huang, Kaiyu; Xie, Lunchen
HeteFed: Heterogeneous Federated Learning with Privacy-Preserving Binary Low-Rank Matrix Decomposition Method Proceedings Article
In: 2023 26th International Conference on Computer Supported Cooperative Work in Design (CSCWD), pp. 1238-1244, 2023.
@inproceedings{10152714,
title = {HeteFed: Heterogeneous Federated Learning with Privacy-Preserving Binary Low-Rank Matrix Decomposition Method},
author = {Jiaqi Liu and Kaiyu Huang and Lunchen Xie},
url = {https://ieeexplore.ieee.org/abstract/document/10152714},
doi = {10.1109/CSCWD57460.2023.10152714},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 26th International Conference on Computer Supported Cooperative Work in Design (CSCWD)},
pages = {1238-1244},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Li, Chaofei; Zhu, Ziyuan; Niu, Ruicheng; Leng, Tao; Meng, Dan
SeAuNet: Semi-Autonomous Encrypted Traffic Classification and Self-labeling Proceedings Article
In: 2023 26th International Conference on Computer Supported Cooperative Work in Design (CSCWD), pp. 1914-1919, 2023.
@inproceedings{10152574,
title = {SeAuNet: Semi-Autonomous Encrypted Traffic Classification and Self-labeling},
author = {Chaofei Li and Ziyuan Zhu and Ruicheng Niu and Tao Leng and Dan Meng},
doi = {10.1109/CSCWD57460.2023.10152574},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 26th International Conference on Computer Supported Cooperative Work in Design (CSCWD)},
pages = {1914-1919},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Shahriari, M; Pardo, D; Kargaran, S; Teijeiro, T
Neural network architecture optimization using automated machine learning for borehole resistivity measurements Journal Article
In: Geophysical Journal International, vol. 234, no. 3, pp. 2488-2501, 2023, ISSN: 0956-540X.
@article{10.1093/gji/ggad249,
title = {Neural network architecture optimization using automated machine learning for borehole resistivity measurements},
author = {M Shahriari and D Pardo and S Kargaran and T Teijeiro},
url = {https://doi.org/10.1093/gji/ggad249},
doi = {10.1093/gji/ggad249},
issn = {0956-540X},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Geophysical Journal International},
volume = {234},
number = {3},
pages = {2488-2501},
abstract = {Deep neural networks (DNNs) offer a real-time solution for the inversion of borehole resistivity measurements to approximate forward and inverse operators. Using extremely large DNNs to approximate the operators is possible, but it demands considerable training time. Moreover, evaluating the network after training also requires a significant amount of memory and processing power. In addition, we may overfit the model. In this work, we propose a scoring function that accounts for the accuracy and size of the DNNs compared to a reference DNNs that provides good approximations for the operators. Using this scoring function, we use DNN architecture search algorithms to obtain a quasi-optimal DNN smaller than the reference network; hence, it requires less computational effort during training and evaluation. The quasi-optimal DNN delivers comparable accuracy to the original large DNN.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Yuan, Xin; Savarese, Pedro; Maire, Michael
Accelerated Training via Incrementally Growing Neural Networks using Variance Transfer and Learning Rate Adaptation Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-12700,
title = {Accelerated Training via Incrementally Growing Neural Networks using Variance Transfer and Learning Rate Adaptation},
author = {Xin Yuan and Pedro Savarese and Michael Maire},
url = {https://doi.org/10.48550/arXiv.2306.12700},
doi = {10.48550/arXiv.2306.12700},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.12700},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Garavagno, Andrea Mattia; Ragusa, Edoardo; Frisoli, Antonio; Gastaldo, Paolo
A hardware-aware neural architecture search algorithm targeting low-end microcontrollers Proceedings Article
In: 2023 18th Conference on Ph.D Research in Microelectronics and Electronics (PRIME), pp. 281-284, 2023.
@inproceedings{10161736,
title = {A hardware-aware neural architecture search algorithm targeting low-end microcontrollers},
author = {Andrea Mattia Garavagno and Edoardo Ragusa and Antonio Frisoli and Paolo Gastaldo},
url = {https://ieeexplore.ieee.org/abstract/document/10161736},
doi = {10.1109/PRIME58259.2023.10161736},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 18th Conference on Ph.D Research in Microelectronics and Electronics (PRIME)},
pages = {281-284},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Qian, Xiaoxue; Liu, Fang; Jiao, Licheng; Zhang, Xiangrong; Huang, Xinyan; Li, Shuo; Chen, Puhua; Liu, Xu
Knowledge transfer evolutionary search for lightweight neural architecture with dynamic inference Journal Article
In: Pattern Recognition, vol. 143, pp. 109790, 2023, ISSN: 0031-3203.
@article{QIAN2023109790,
title = {Knowledge transfer evolutionary search for lightweight neural architecture with dynamic inference},
author = {Xiaoxue Qian and Fang Liu and Licheng Jiao and Xiangrong Zhang and Xinyan Huang and Shuo Li and Puhua Chen and Xu Liu},
url = {https://www.sciencedirect.com/science/article/pii/S0031320323004880},
doi = {https://doi.org/10.1016/j.patcog.2023.109790},
issn = {0031-3203},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Pattern Recognition},
volume = {143},
pages = {109790},
abstract = {Relying on the availability of massive labeled samples, most neural architecture search (NAS) methods focus on searching large and complex models; and adopt fixed structures and parameters at the inference stage. Few approaches automatically design lightweight networks for label-limited tasks and further consider the inference differences between inputs. To address these issues, we introduce evolutionary computation (EC) and attention mechanism and propose a knowledge transfer evolutionary search for lightweight neural architecture with dynamic inference, then verify it using synthetic aperture radar (SAR) images. SAR image classification is a typical label-limited task due to the inherent imaging mechanism of SAR. We design the EC-based architecture search and attention-based dynamic inference for SAR image scene classification. Specifically, we build a SAR-tailored search space, explore topology pruning-based mutation operators to search lightweight architectures, and further design a dynamic Ridgelet convolution capable of adaptive reasoning to enhance the representation ability of searched lightweight networks. Moreover, we propose a knowledge transfer training strategy and hybrid evaluation criteria to ensure searching quickly and robustly. Experimental results show that the proposed method can search for superior neural architectures, thus improving the classification performance of SAR images.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Wang, Qianchao; Kapuza, Itamar; Baimel, Dmitry; Belikov, Juri; Levron, Yoash; Machlev, Ram
Neural Architecture Search (NAS) for designing optimal power quality disturbance classifiers Journal Article
In: Electric Power Systems Research, vol. 223, pp. 109574, 2023, ISSN: 0378-7796.
@article{WANG2023109574,
title = {Neural Architecture Search (NAS) for designing optimal power quality disturbance classifiers},
author = {Qianchao Wang and Itamar Kapuza and Dmitry Baimel and Juri Belikov and Yoash Levron and Ram Machlev},
url = {https://www.sciencedirect.com/science/article/pii/S0378779623004637},
doi = {https://doi.org/10.1016/j.epsr.2023.109574},
issn = {0378-7796},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Electric Power Systems Research},
volume = {223},
pages = {109574},
abstract = {Deep learning techniques have recently demonstrated outstanding success when used for Power Quality Disturbance (PQD) classification. However, a core obstacle is that deep neural networks (DNN)s are complex models, and their architecture is designed using trial and error processes. Accordingly, the problem of finding the optimal architecture can be considered as a problem that consists of high-dimensional solutions. Meanwhile, in the last couple of years, Neural Architecture Search (NAS) techniques have been developed to efficiently find the best possible performance architecture for a specific task. In this light, the goal of this research is to develop a method to find optimal PQD classifiers using the NAS technique, based on an evolutionary algorithm. This method can converge efficiently to an optimal DNN architecture. Thus, a classifier that achieves high accuracy for PQDs classification is provided using limited resources and with minimal human intervention. This idea is demonstrated on two different DNN typologies—convolutional neural networks (CNN) and Bi-directional long short-term memory (Bi-LSTM). By adopting this method, the results of the generated PQD classifiers are more accurate when compared to recently developed classifiers.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Savchenko, Andrey V.; Savchenko, Lyudmila V.; Makarov, Ilya
Fast Search of Face Recognition Model for a Mobile Device Based on Neural Architecture Comparator Journal Article
In: IEEE Access, vol. 11, pp. 65977-65990, 2023.
@article{10168107,
title = {Fast Search of Face Recognition Model for a Mobile Device Based on Neural Architecture Comparator},
author = {Andrey V. Savchenko and Lyudmila V. Savchenko and Ilya Makarov},
url = {https://ieeexplore.ieee.org/document/10168107},
doi = {10.1109/ACCESS.2023.3290902},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Access},
volume = {11},
pages = {65977-65990},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Shi, Jiachen; Liu, Wenzhen; Zhou, Guoqiang; Zhou, Yuming
AutoInfo GAN: Toward a better image synthesis GAN framework for high-fidelity few-shot datasets via NAS and contrastive learning Journal Article
In: Knowledge-Based Systems, vol. 276, pp. 110757, 2023, ISSN: 0950-7051.
@article{SHI2023110757,
title = {AutoInfo GAN: Toward a better image synthesis GAN framework for high-fidelity few-shot datasets via NAS and contrastive learning},
author = {Jiachen Shi and Wenzhen Liu and Guoqiang Zhou and Yuming Zhou},
url = {https://www.sciencedirect.com/science/article/pii/S0950705123005075},
doi = {https://doi.org/10.1016/j.knosys.2023.110757},
issn = {0950-7051},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Knowledge-Based Systems},
volume = {276},
pages = {110757},
abstract = {Background:
Generative adversarial networks (GANs) are vital techniques for synthesizing high-fidelity images. Recent studies have applied them to generation tasks under small-data scenarios. Most studies do not directly train GANs on few-shot datasets, which have small data samples; instead, they borrow methods to transfer knowledge from large datasets to GANs with small ones. Partial fine-tuning of GANs is difficult to ensure the transfer performance, especially when the image domains are of great difference. FastGAN firstly trains GAN with small data samples by a carefully designed skip layer exception (SLE) connection to improve synthesis and an unsupervised discriminator to avoid overfitting. Problem. However, in FastGAN, different designs of SLE connections and operation settings would lead to great differences in synthesis performance. It is necessary to find the most appropriate ways of architecture design. Meanwhile, FastGAN merely improves discriminator learning, but ignores that the generator learning process is also insufficient due to small data samples.
Objective:
Based on FastGAN, this study aims to find the best generator designs and then improve the training process of it via unsupervised learning. Methods. This work applies a reinforcement learning neural architecture search method to find the optimal GAN architecture and an unsupervised contrastive loss function assisted by a discriminator to optimize generator learning. These two methods constitute our AutoInfoGAN.
Results:
Experiments were conducted on 11 datasets using AutoInfoGAN, covering a wide range of image domains, achieving better results than state-of-the-art (SOTA) models.
Conclusion:
The experimental results demonstrate the SOTA performance of our proposed AutoInfo GAN on few-shot datasets, and we are cautioned that although instance normalization (IN) improves synthesized image quality, it performed poorly in our mode-collapse test.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Generative adversarial networks (GANs) are vital techniques for synthesizing high-fidelity images. Recent studies have applied them to generation tasks under small-data scenarios. Most studies do not directly train GANs on few-shot datasets, which have small data samples; instead, they borrow methods to transfer knowledge from large datasets to GANs with small ones. Partial fine-tuning of GANs is difficult to ensure the transfer performance, especially when the image domains are of great difference. FastGAN firstly trains GAN with small data samples by a carefully designed skip layer exception (SLE) connection to improve synthesis and an unsupervised discriminator to avoid overfitting. Problem. However, in FastGAN, different designs of SLE connections and operation settings would lead to great differences in synthesis performance. It is necessary to find the most appropriate ways of architecture design. Meanwhile, FastGAN merely improves discriminator learning, but ignores that the generator learning process is also insufficient due to small data samples.
Objective:
Based on FastGAN, this study aims to find the best generator designs and then improve the training process of it via unsupervised learning. Methods. This work applies a reinforcement learning neural architecture search method to find the optimal GAN architecture and an unsupervised contrastive loss function assisted by a discriminator to optimize generator learning. These two methods constitute our AutoInfoGAN.
Results:
Experiments were conducted on 11 datasets using AutoInfoGAN, covering a wide range of image domains, achieving better results than state-of-the-art (SOTA) models.
Conclusion:
The experimental results demonstrate the SOTA performance of our proposed AutoInfo GAN on few-shot datasets, and we are cautioned that although instance normalization (IN) improves synthesized image quality, it performed poorly in our mode-collapse test.
Ball, Richard; Kruger, Hennie; Drevin, Lynette
Anomaly detection using autoencoders with network analysis features Journal Article
In: ORiON, vol. 39, pp. 1-44, 2023.
@article{articlec,
title = {Anomaly detection using autoencoders with network analysis features},
author = {Richard Ball and Hennie Kruger and Lynette Drevin},
url = {https://www.researchgate.net/publication/371959893_Anomaly_detection_using_autoencoders_with_network_analysis_features},
doi = {10.5784/39-1-711},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {ORiON},
volume = {39},
pages = {1-44},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Sarti, Simone; Lomurno, Eugenio; Matteucci, Matteo
Neural Architecture Transfer 2: A Paradigm for Improving Efficiency in Multi-Objective Neural Architecture Search Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-00960,
title = {Neural Architecture Transfer 2: A Paradigm for Improving Efficiency in Multi-Objective Neural Architecture Search},
author = {Simone Sarti and Eugenio Lomurno and Matteo Matteucci},
url = {https://doi.org/10.48550/arXiv.2307.00960},
doi = {10.48550/arXiv.2307.00960},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.00960},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Wang, Ziqing; Zhao, Qidong; Cui, Jinku; Liu, Xu; Xu, Dongkuan
AutoST: Training-free Neural Architecture Search for Spiking Transformers Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-00293,
title = {AutoST: Training-free Neural Architecture Search for Spiking Transformers},
author = {Ziqing Wang and Qidong Zhao and Jinku Cui and Xu Liu and Dongkuan Xu},
url = {https://doi.org/10.48550/arXiv.2307.00293},
doi = {10.48550/arXiv.2307.00293},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.00293},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Subbicini, Giorgia; Lavagno, Luciano; Lazarescu, Mihai T.
Enhanced Exploration of Neural Network Models for Indoor Human Monitoring Proceedings Article
In: 2023 9th International Workshop on Advances in Sensors and Interfaces (IWASI), pp. 109-114, 2023.
@inproceedings{10164436,
title = {Enhanced Exploration of Neural Network Models for Indoor Human Monitoring},
author = {Giorgia Subbicini and Luciano Lavagno and Mihai T. Lazarescu},
url = {https://ieeexplore.ieee.org/abstract/document/10164436},
doi = {10.1109/IWASI58316.2023.10164436},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 9th International Workshop on Advances in Sensors and Interfaces (IWASI)},
pages = {109-114},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Li, Guihong; Hoang, Duc; Bhardwaj, Kartikeya; Lin, Ming; Wang, Zhangyang; Marculescu, Radu
Zero-Shot Neural Architecture Search: Challenges, Solutions, and Opportunities Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-01998,
title = {Zero-Shot Neural Architecture Search: Challenges, Solutions, and Opportunities},
author = {Guihong Li and Duc Hoang and Kartikeya Bhardwaj and Ming Lin and Zhangyang Wang and Radu Marculescu},
url = {https://doi.org/10.48550/arXiv.2307.01998},
doi = {10.48550/arXiv.2307.01998},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.01998},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Du, Wentian; Chen, Jie; Zhang, Chaochen; Zhao, Po; Wan, Huiyao; Zhou, Zheng; Cao, Yice; Huang, Zhixiang; Li, Yingsong; Wu, Bocai
SARNas: A Hardware-Aware SAR Target Detection Algorithm via Multiobjective Neural Architecture Search Journal Article
In: IEEE Transactions on Geoscience and Remote Sensing, pp. 1-1, 2023.
@article{10173551,
title = {SARNas: A Hardware-Aware SAR Target Detection Algorithm via Multiobjective Neural Architecture Search},
author = {Wentian Du and Jie Chen and Chaochen Zhang and Po Zhao and Huiyao Wan and Zheng Zhou and Yice Cao and Zhixiang Huang and Yingsong Li and Bocai Wu},
doi = {10.1109/TGRS.2023.3292618},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Luo, Jianxiang; Hu, Junyi; Pang, Tianji; Huang, Weihao; Liu, Chuang
Dynamical Isometry based Rigorous Fair Neural Architecture Search Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-02263,
title = {Dynamical Isometry based Rigorous Fair Neural Architecture Search},
author = {Jianxiang Luo and Junyi Hu and Tianji Pang and Weihao Huang and Chuang Liu},
url = {https://doi.org/10.48550/arXiv.2307.02263},
doi = {10.48550/arXiv.2307.02263},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.02263},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Shang, Ronghua; Zhu, Songling; Liu, Hangcheng; Ma, Teng; Zhang, Weitong; Feng, Jie; Jiao, Licheng; Stolkin, Rustam
Evolutionary architecture search via adaptive parameter control and gene potential contribution Journal Article
In: Swarm and Evolutionary Computation, vol. 82, pp. 101354, 2023, ISSN: 2210-6502.
@article{SHANG2023101354,
title = {Evolutionary architecture search via adaptive parameter control and gene potential contribution},
author = {Ronghua Shang and Songling Zhu and Hangcheng Liu and Teng Ma and Weitong Zhang and Jie Feng and Licheng Jiao and Rustam Stolkin},
url = {https://www.sciencedirect.com/science/article/pii/S221065022300127X},
doi = {https://doi.org/10.1016/j.swevo.2023.101354},
issn = {2210-6502},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Swarm and Evolutionary Computation},
volume = {82},
pages = {101354},
abstract = {Evolutionary neural architecture search (ENAS) algorithms benefit from the non-convex optimization capability of evolutionary computation. However, most ENAS algorithms use a genetic algorithm (GA) with fixed parameter settings, which limits the algorithm’s search performance. Furthermore, the information generated during the evolutionary process is often ignored, while this information can be helpful for guiding the evolutionary direction of the population. This paper proposes a novel ENAS algorithm based on adaptive parameter control and gene potential contribution (AG-ENAS) to evolve neural networks efficiently. Firstly, an adaptive parameter adjustment mechanism is designed, based on population diversity and fitness. This enables better-informed adaptation of related parameters of genetic operators. Secondly, the mutation operator guided by the gene potential contribution of genes tends to produce better offspring. The gene potential contribution reflects the positive effect of the current gene on fitness. It guides the evolution by weighting the more valuable genes with the distribution index matrix. Finally, the concept of aging is introduced into the environmental selection, to offer more opportunities to the young generation and alleviate premature convergence. The proposed algorithm has been evaluated on eight different datasets, and compared against 44 state-of-the-art algorithms from the literature. The experimental results show that the network designed by AG-ENAS obtains higher classification accuracy than manual-designed networks such as SENet, DenseNet, and other ENAS algorithms such as Large-Evo and AE-CNN.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Chen, Bozhou; Wang, Hongzhi
SoftStep relaxation for mining optimal convolution kernel Journal Article
In: Knowledge-Based Systems, vol. 276, pp. 110755, 2023, ISSN: 0950-7051.
@article{CHEN2023110755,
title = {SoftStep relaxation for mining optimal convolution kernel},
author = {Bozhou Chen and Hongzhi Wang},
url = {https://www.sciencedirect.com/science/article/pii/S0950705123005051},
doi = {https://doi.org/10.1016/j.knosys.2023.110755},
issn = {0950-7051},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Knowledge-Based Systems},
volume = {276},
pages = {110755},
abstract = {An efficient convolution kernel could transform images into more expressive representations, which usually determines the quality of image classification models. Algorithms in the field of hyper-parameter optimization and neural architecture search could be applied to mining optimal kernels, while hyper-parameter optimization algorithms always require multiple training sessions to evaluate hyper-parameter configurations, resulting in a high time consumption. Recent gradient-based architecture search algorithms could deal with such problems within a single training session. However, there are currently no gradient-based NAS algorithms that could be used to efficiently optimize convolution kernels. This paper proposes a novel gradient-based algorithm, SoftStep, which could precisely and efficiently fine-tune the hyper-parameters of kernels. The proposed SoftStep is applied for the optimization of the convolution kernel’s key hyper-parameters and generalized to more hyper-parameters in deep learning. In experiments, architecture search tasks on multiple challenging datasets are compared to various state-of-art hyper-parameter optimization and architecture search algorithms. The reported results show that SoftStep could mine optimal kernels far more efficiently than other SOTA methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An, Xiaojie; Ma, Lianbo; Zhou, Yuee; Li, Nan; Xing, Tiejun; Wen, Yingyou; Liu, Chang; Shi, Haibo
Neural Architecture Search Based on Improved Brain Storm Optimization Algorithm Proceedings Article
In: Tan, Ying; Shi, Yuhui; Luo, Wenjian (Ed.): Advances in Swarm Intelligence, pp. 334–344, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-36622-2.
@inproceedings{10.1007/978-3-031-36622-2_27,
title = {Neural Architecture Search Based on Improved Brain Storm Optimization Algorithm},
author = {Xiaojie An and Lianbo Ma and Yuee Zhou and Nan Li and Tiejun Xing and Yingyou Wen and Chang Liu and Haibo Shi},
editor = {Ying Tan and Yuhui Shi and Wenjian Luo},
url = {https://link.springer.com/chapter/10.1007/978-3-031-36622-2_27},
isbn = {978-3-031-36622-2},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Advances in Swarm Intelligence},
pages = {334–344},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {The performance of deep neural networks (DNNs) often depends on the design of their architectures. But designing a DNN with good performance is a difficult and knowledge-intensive process. In this paper, we propose a neural architecture search method based on improved brain storm optimization (BSO) algorithm to efficiently deal with image classification tasks. BSO successfully transposes the human brainstorming process to design of optimization algorithms, which typically uses grouping, substitution, and creation operators to generate as many solutions as possible to approach the global optimization of the problem generation by generation. However, the BSO algorithm using clustering methods for grouping increases the computational burden, so we use the BSO algorithm based on simple grouping methods to solve the optimal architecture of the neural architecture search (NAS). We also redesigned the search space and designed an efficient encoding strategy for each individual.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Gopal, Bhavna; Sridhar, Arjun; Zhang, Tunhou; Chen, Yiran
LISSNAS: Locality-based Iterative Search Space Shrinkage for Neural Architecture Search Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-03110,
title = {LISSNAS: Locality-based Iterative Search Space Shrinkage for Neural Architecture Search},
author = {Bhavna Gopal and Arjun Sridhar and Tunhou Zhang and Yiran Chen},
url = {https://doi.org/10.48550/arXiv.2307.03110},
doi = {10.48550/arXiv.2307.03110},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.03110},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Zhang, Kehao; Jin, Huaiping; Jin, Huaikang; Wang, Bin; Yu, Wangyang
Gated Recurrent Unit Neural Networks for Wind Power Forecasting based on Surrogate-Assisted Evolutionary Neural Architecture Search Proceedings Article
In: 2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS), pp. 1774-1779, 2023.
@inproceedings{10166074,
title = {Gated Recurrent Unit Neural Networks for Wind Power Forecasting based on Surrogate-Assisted Evolutionary Neural Architecture Search},
author = {Kehao Zhang and Huaiping Jin and Huaikang Jin and Bin Wang and Wangyang Yu},
url = {https://ieeexplore.ieee.org/abstract/document/10166074},
doi = {10.1109/DDCLS58216.2023.10166074},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)},
pages = {1774-1779},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Gong, Maoguo; Gao, Tianqi; Zhang, Mingyang; Li, Wei; Wang, Zhibin; Li, Dezhong
An M-Nary SAR Image Change Detection Based on GAN Architecture Search Journal Article
In: IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1-18, 2023.
@article{10175564,
title = {An M-Nary SAR Image Change Detection Based on GAN Architecture Search},
author = {Maoguo Gong and Tianqi Gao and Mingyang Zhang and Wei Li and Zhibin Wang and Dezhong Li},
url = {https://ieeexplore.ieee.org/abstract/document/10175564},
doi = {10.1109/TGRS.2023.3293190},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
volume = {61},
pages = {1-18},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Chen, Shengbo; Chen, Shijie; Lei, Zhou
FAS-reid: Fair Architecture Search for Person Re-IDentification Proceedings Article
In: 2023 IEEE 3rd International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA), pp. 544-551, 2023.
@inproceedings{10165411,
title = {FAS-reid: Fair Architecture Search for Person Re-IDentification},
author = {Shengbo Chen and Shijie Chen and Zhou Lei},
url = {https://ieeexplore.ieee.org/abstract/document/10165411},
doi = {10.1109/ICIBA56860.2023.10165411},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE 3rd International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA)},
volume = {3},
pages = {544-551},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Qin, Haiyan; Zeng, Yejun; Bai, Jinyu; Kang, Wang
Searching Tiny Neural Networks for Deployment on Embedded FPGA Proceedings Article
In: 2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS), pp. 1-5, 2023.
@inproceedings{10168571,
title = {Searching Tiny Neural Networks for Deployment on Embedded FPGA},
author = {Haiyan Qin and Yejun Zeng and Jinyu Bai and Wang Kang},
doi = {10.1109/AICAS57966.2023.10168571},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS)},
pages = {1-5},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Deng, Shuchao; Lv, Zeqiong; Galván, Edgar; Sun, Yanan
Evolutionary Neural Architecture Search for Facial Expression Recognition Journal Article
In: IEEE Transactions on Emerging Topics in Computational Intelligence, pp. 1-15, 2023.
@article{10177272,
title = {Evolutionary Neural Architecture Search for Facial Expression Recognition},
author = {Shuchao Deng and Zeqiong Lv and Edgar Galván and Yanan Sun},
doi = {10.1109/TETCI.2023.3289974},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Transactions on Emerging Topics in Computational Intelligence},
pages = {1-15},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Zhao, Yiyang; Guo, Tian
Carbon-Efficient Neural Architecture Search Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-04131,
title = {Carbon-Efficient Neural Architecture Search},
author = {Yiyang Zhao and Tian Guo},
url = {https://doi.org/10.48550/arXiv.2307.04131},
doi = {10.48550/arXiv.2307.04131},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.04131},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Yang, Shangshang; Ma, Haiping; Zhen, Cheng; Tian, Ye; Zhang, Limiao; Jin, Yaochu; Zhang, Xingyi
Designing Novel Cognitive Diagnosis Models via Evolutionary Multi-Objective Neural Architecture Search Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-04429,
title = {Designing Novel Cognitive Diagnosis Models via Evolutionary Multi-Objective Neural Architecture Search},
author = {Shangshang Yang and Haiping Ma and Cheng Zhen and Ye Tian and Limiao Zhang and Yaochu Jin and Xingyi Zhang},
url = {https://doi.org/10.48550/arXiv.2307.04429},
doi = {10.48550/arXiv.2307.04429},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.04429},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Dutta, Oshin; Kanvar, Tanu; Agarwal, Sumeet
Search-time Efficient Device Constraints-Aware Neural Architecture Search Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2307-04443,
title = {Search-time Efficient Device Constraints-Aware Neural Architecture Search},
author = {Oshin Dutta and Tanu Kanvar and Sumeet Agarwal},
url = {https://doi.org/10.48550/arXiv.2307.04443},
doi = {10.48550/arXiv.2307.04443},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2307.04443},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
Wang, Yanxin; Yan, Jing; Qi, Meirong; Wang, Jianhua; Geng, Yingsan
A Novel Meta-Learning and Network Architecture Search Approach for Few-Shot High-Voltage Circuit Breaker Fault Diagnosis Proceedings Article
In: 2023 IEEE 6th International Electrical and Energy Conference (CIEEC), pp. 122-127, 2023.
@inproceedings{10166467,
title = {A Novel Meta-Learning and Network Architecture Search Approach for Few-Shot High-Voltage Circuit Breaker Fault Diagnosis},
author = {Yanxin Wang and Jing Yan and Meirong Qi and Jianhua Wang and Yingsan Geng},
url = {https://ieeexplore.ieee.org/abstract/document/10166467},
doi = {10.1109/CIEEC58067.2023.10166467},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE 6th International Electrical and Energy Conference (CIEEC)},
pages = {122-127},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Anthony, Quentin; Xu, Lang; Shafi, Aamir; Subramoni, Hari; Panda, Dhabaleswar K. DK
ScaMP: Scalable Meta-Parallelism for Deep Learning Search Proceedings Article
In: 2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing (CCGrid), pp. 391-402, 2023.
@inproceedings{10171481,
title = {ScaMP: Scalable Meta-Parallelism for Deep Learning Search},
author = {Quentin Anthony and Lang Xu and Aamir Shafi and Hari Subramoni and Dhabaleswar K. DK Panda},
doi = {10.1109/CCGrid57682.2023.00044},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing (CCGrid)},
pages = {391-402},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}