Mystique: User-Level Adaptation for Real-Time Video Analytics in Edge Networks via Meta-RL
- 9 December 2024
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (5) , 3615-3632
Abstract
Deep neural network (DNN)-based real-time video analytics service, as a core module for numerous crucial applications such as augmented reality (AR), has garnered increasing research attention, where mobile edge computing (MEC) is often leveraged to mitigate its real-time processing burden on resource-constrained user devices. For Quality of Experience (QoE) optimization, latest works employ reinforcement learning (RL)-based methods to adaptively adjust configurations (e.g., resolution and frame rate), yet still presenting significant challenges. Firstly, we observe a substantial diversity in QoE patterns among users. Given that existing methods integrate a fixed QoE pattern in parameter training, it is intuitive to customize a policy network for each user. However, this necessitates significant training investment, failing to support on-the-fly deployment for new users. Secondly, given the dual dynamics from both the network and video content in edge video analytics system, existing methods often fall into the dilemma of fitting newly emerged and diverse system states with offline-trained fixed parameters. While it is promising to employ online learning algorithms, most of them struggle to catch up with the high dynamics. We hence propose Mystique. In real-time edge video analytics domain, it is the first meta-RL-based user-level configuration adaptation framework. Mystique establishes an initial model in offline meta training with model-agnostic meta-learning (MAML), enabling swift online adaptation to new users and system states through limited gradient updates from initial parameters. Comprehensive experiments illustrate that Mystique can improve QoE by 42% on average compared to prior works.Keywords
Funding Information
- Science and Technology Project of State Grid Company, LTD
- Research on data aggregation and dynamic interaction technology of enterprise-level real-time measurement data center (5108-202218280A-2-399-XG)
- Nanjing University-China Mobile Communications Group Company, Ltd
This publication has 39 references indexed in Scilit:
- Ego4D: Around the World in 3,000 Hours of Egocentric Video2022
- Meta-Learning with Task-Adaptive Loss Function for Few-Shot Learning2021
- Deepmpc: A Mixture Abr Approach Via Deep Learning And Mpc2020
- Edge Assisted Mobile Semantic Visual SLAM2020
- FastVA: Deep Learning Video Analytics Through Edge Processing and NPU in Mobile2020
- Joint Configuration Adaptation and Bandwidth Allocation for Edge-based Real-time Video Analytics2020
- Hetero-Edge: Orchestration of Real-time Vision Applications on Heterogeneous Edge Clouds2019
- An Edge Network Orchestrator for Mobile Augmented Reality2018
- Focal Loss for Dense Object Detection2017
- You Only Look Once: Unified, Real-Time Object Detection2016