ETNAS: An energy consumption task-driven neural architecture search

被引:1
|
作者
Dong, Dong [1 ,2 ]
Jiang, Hongxu [1 ,2 ]
Wei, Xuekai [3 ]
Song, Yanfei [1 ,2 ]
Zhuang, Xu [4 ]
Wang, Jason [5 ]
机构
[1] Beihang Univ, Beijing Key Lab Digital Media, State Key Lab Virtual Real Technol & Syst, Beijing 100191, Peoples R China
[2] BeiHang Univ, Hangzhou Innovat Inst, Hangzhou 310051, Peoples R China
[3] Chongqing Univ, Sch Comp Sci, Chongqing 400044, Peoples R China
[4] Guangdong Opel Mobile Commun Co Ltd, OPPO, Chengdu 610000, Peoples R China
[5] Guangdong Opel Mobile Commun Co Ltd, OPPO, Nanjing 210000, Peoples R China
来源
SUSTAINABLE COMPUTING-INFORMATICS & SYSTEMS | 2023年 / 40卷
关键词
Neural architecture search; Differentiable architecture search; Power consumption; Task-driven; Multi-objective optimization;
D O I
10.1016/j.suscom.2023.100926
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Neural Architecture Search (NAS) is crucial in the field of sustainable computing as it facilitates the development of highly efficient and effective neural networks. However, it cannot automate the deployment of neural networks to accommodate specific hardware resources and task requirements. This paper introduces ETNAS, which is a hardware-aware multi-objective optimal neural network architecture search algorithm based on the differentiable neural network architecture search method (DARTS). The algorithm searches for a lower-power neural network architecture with guaranteed inference accuracy by modifying the loss function of the differentiable neural network architecture search. We modify the dense network in DARTS to simultaneously search for networks with a lower memory footprint, enabling them to run on memory-constrained edge-end devices. We collected data on the power consumption and time consumption of numerous common operators on FPGA and Domain-Specific Architectures (DSA). The experimental results demonstrate that ETNAS achieves comparable accuracy performance and time efficiency while consuming less power compared to state-of-the-art algorithms, thereby validating its effectiveness in practical applications and contributing to the reduction of carbon emissions in intelligent cyber-physical systems (ICPS) edge computing inference.
引用
收藏
页数:7
相关论文
共 50 条
  • [21] SOCIAL INTELLIGENCE FOR A TASK-DRIVEN ASSISTIVE ROBOT
    Nejat, Goldie
    Ficocelli, Maurizio
    PROCEEDINGS OF THE ASME INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE 2010, VOL 2, PTS A AND B, 2010, : 1555 - 1563
  • [22] Meta Learning for Task-Driven Video Summarization
    Li, Xuelong
    Li, Hongli
    Dong, Yongsheng
    IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2020, 67 (07) : 5778 - 5786
  • [23] Guided model creation: A task-driven approach
    Lahtinen, Samuel
    Peltonen, Jari
    Hammouda, Imed
    Koskimies, Kai
    IEEE SYMPOSIUM ON VISUAL LANGUAGES AND HUMAN-CENTRIC COMPUTING, PROCEEDINGS, 2006, : 89 - +
  • [24] Quantization in task-driven sensing and distributed processing
    Gray, Robert M.
    2006 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-13, 2006, : 5907 - 5910
  • [25] Task-driven camera operations for robotic exploration
    Hughes, SB
    Lewis, M
    IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART A-SYSTEMS AND HUMANS, 2005, 35 (04): : 513 - 522
  • [26] A Task-Driven Feedback Imager with Uncertainty Driven Hybrid Control
    Mudassar, Burhan A.
    Saha, Priyabrata
    Wolf, Marilyn
    Mukhopadhyay, Saibal
    SENSORS, 2021, 21 (08)
  • [27] Task-Driven Prompt Evolution for Foundation Models
    Sathish, Rachana
    Venkataramani, Rahul
    Shriram, K. S.
    Sudhakar, Prasad
    MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2023 WORKSHOPS, 2023, 14393 : 256 - 264
  • [28] Selfish Task-Driven Routing in Hybrid Networks
    Li, Yupeng
    Tan, Haisheng
    Wang, Yongcai
    Han, Zhenhua
    Lau, Francis C. M.
    2015 13th International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt), 2015, : 387 - 394
  • [29] Task-Driven Composition of Web User Interfaces
    Betermieux, Stefan
    Bomsdorf, Birgit
    COMPUTER-AIDED DESIGN OF USER INTERFACES VI, 2009, : 233 - +
  • [30] Task-Driven Variability Model for Speaker Verification
    Chen Chen
    Jiqing Han
    Circuits, Systems, and Signal Processing, 2020, 39 : 3125 - 3144