Edge-computing-assisted intelligent processing of AI-generated image content

被引:0
|
作者
Suzhen Wang
Yongchen Deng
Lisha Hu
Ning Cao
机构
[1] Hebei University of Economics and Business,School of Information Technology
[2] Wuxi Vocational College of Science and Technology,School of Integrated Circuits
[3] Shandong Vocational and Technical University of International Studies,School of Information Engineering
来源
关键词
AIGIC; Edge computing; Serverless computing; WASM; E-PPO2;
D O I
暂无
中图分类号
学科分类号
摘要
Artificial intelligence-generated image content (AIGIC) is produced through the extraction of features and patterns from a vast image dataset, requiring substantial computational resources for training. Due to the limited computational resources of terminal devices, efficiently processing and responding to AIGIC has emerged as a critical concern in current research. To address this challenge, the present paper proposed the utilization of edge computing technology. Edge computing enables the offloading of certain training tasks to edge nodes, facilitating expedited task offloading strategies that empower terminal devices to generate image content efficiently. Building upon the edge serverless architecture, this paper introduces an edge serverless computing framework based on Web Assembly (WASM). Notably, this framework effectively resolves the latency issue associated with container cold start in serverless computing. Additionally, to enhance the collaborative capabilities of edge nodes, entropy-based Proximal Policy Optimization (E-PPO2) is proposed. This algorithm enables each edge node to share global rewards, continually update parameters, and ultimately derive the optimal response strategy, thereby harnessing edge device resources to their fullest potential. Finally, the efficacy of the proposed serverless computing architecture based on WASM is demonstrated through the evaluation of 13 benchmark functions. Comparative analyses with four task offloading algorithms highlight that the E-PPO2 algorithm, proposed in this article, significantly reduces task execution latency, facilitating rapid processing and response in AIGIC scenarios.
引用
收藏
相关论文
共 50 条
  • [31] Effect of disclosing AI-generated content on prosocial advertising evaluation
    Baek, Tae Hyun
    Kim, Jungkeun
    Kim, Jeong Hyun
    INTERNATIONAL JOURNAL OF ADVERTISING, 2024,
  • [32] 'AI-navigating' or 'AI-sinking'? An analysis of verbs in research articles titles suspicious of containing AI-generated/assisted content
    Comas-Forgas, Ruben
    Koulouris, Alexandros
    Kouis, Dimitris
    LEARNED PUBLISHING, 2025, 38 (01)
  • [33] Workshop: Using AI-Generated Content to Support the Writing Process
    Vance, Bremen
    Brewer, Pam Estes
    Duin, Ann Hill
    2023 IEEE INTERNATIONAL PROFESSIONAL COMMUNICATION CONFERENCE, PROCOMM, 2023, : 168 - 170
  • [34] Examining the role of compression in influencing AI-generated image authenticity
    Xiaohan Fang
    Peilin Chen
    Meng Wang
    Shiqi Wang
    Scientific Reports, 15 (1)
  • [35] Enhancing Interpretability in AI-Generated Image Detection with Genetic Programming
    Lin, Mingqian
    Shang, Lin
    Gao, Xiaoying
    2023 23RD IEEE INTERNATIONAL CONFERENCE ON DATA MINING WORKSHOPS, ICDMW 2023, 2023, : 371 - 378
  • [36] Beijing Internet Court recognizes copyright in AI-generated image
    Wen, Tingting
    JOURNAL OF INTELLECTUAL PROPERTY LAW & PRACTICE, 2024, 19 (03) : 250 - 255
  • [37] Digital tourism interpretation content quality: A comparison between AI-generated content and professional-generated content
    Zhang, Jiahua Jarrett
    Wang, Ying Wendy
    Ruan, Qian
    Yang, Yang
    TOURISM MANAGEMENT PERSPECTIVES, 2024, 53
  • [38] THE ARTIFICIAL INTELLIGENCE BEHIND THE PODCAST MICROPHONE: AI-GENERATED CONTENT IN PODCASTING
    Vrzalikova, Katerina
    MARKETING AND MEDIA IDENTITY: AI-THE FUTURE OF TODAY, 2023, 2023, : 437 - 445
  • [39] AI-Generated Content, Creative Freelance Work and Hospitality and Tourism Marketing
    Tuomi, Aarni
    INFORMATION AND COMMUNICATION TECHNOLOGIES IN TOURISM 2023, ENTER 2023, 2023, : 323 - 328
  • [40] Online AI-Generated Content Request Scheduling with Deep Reinforcement Learning
    Feng, Chenglong
    Zheng, Ying
    Xu, Yuedong
    IEEE INFOCOM 2024-IEEE CONFERENCE ON COMPUTER COMMUNICATIONS WORKSHOPS, INFOCOM WKSHPS 2024, 2024,