Comyco: Quality-Aware Adaptive Video Streaming via Imitation Learning

被引:66
作者
Huang, Tianchi [1 ]
Zhou, Chao [2 ]
Zhang, Rui-Xiao [1 ]
Wu, Chenglei [1 ]
Yao, Xin [1 ]
Sun, Lifeng [3 ]
机构
[1] Tsinghua Univ, Dept Comp Sci & Technol, Beijing, Peoples R China
[2] Beijing Kuaishou Technol Co Ltd, Beijing, Peoples R China
[3] Tsinghua Univ, Dept Comp Sci & Technol, BNRist, Beijing, Peoples R China
来源
PROCEEDINGS OF THE 27TH ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA (MM'19) | 2019年
基金
国家重点研发计划;
关键词
Imitation Learning; Quality-aware; Adaptive Video Streaming;
D O I
10.1145/3343031.3351014
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Learning-based Adaptive Bit Rate (ABR) method, aiming to learn outstanding strategies without any presumptions, has become one of the research hotspots for adaptive streaming. However, it is still suffering from several issues, i.e., low sample efficiency and lack of awareness of the video quality information. In this paper, we propose Comyco, a video quality-aware ABR approach that enormously improves the learning-based methods by tackling the above issues. Comyco trains the policy via imitating expert trajectories given by the instant solver, which can not only avoid redundant exploration but also make better use of the collected samples. Meanwhile, Comyco attempts to pick the chunk with higher perceptual video qualities rather than video bitrates. To achieve this, we construct Comyco's neural network architecture, video datasets and QoE metrics with video quality features. Using trace-driven and real world experiments, we demonstrate significant improvements of Comyco's sample efficiency in comparison to prior work, with 1700x improvements in terms of the number of samples required and 16x improvements on training time required. Moreover, results illustrate that Comyco outperforms previously proposed methods, with the improvements on average QoE of 7.5% - 16.79%. Especially, Comyco also surpasses state-of-the-art approach Pensieve by 7.37% on average video quality under the same rebuffering time.
引用
收藏
页码:429 / 437
页数:9
相关论文
共 45 条
  • [1] Aaron A, 2015, IEEE IMAGE PROC, P1732, DOI 10.1109/ICIP.2015.7351097
  • [2] Abadi M, 2016, PROCEEDINGS OF OSDI'16: 12TH USENIX SYMPOSIUM ON OPERATING SYSTEMS DESIGN AND IMPLEMENTATION, P265
  • [3] Abar T, 2017, INT WIREL COMMUN, P1395, DOI 10.1109/IWCMC.2017.7986488
  • [4] Oboe: Auto-tuning Video ABR Algorithms to Network Conditions
    Akhtar, Zahaib
    Nam, Yun Seong
    Govindan, Ramesh
    Rao, Sanjay
    Chen, Jessica
    Katz-Bassett, Ethan
    Ribeiro, Bruno
    Zhan, Jibin
    Zhang, Hui
    [J]. PROCEEDINGS OF THE 2018 CONFERENCE OF THE ACM SPECIAL INTEREST GROUP ON DATA COMMUNICATION (SIGCOMM '18), 2018, : 44 - 58
  • [5] [Anonymous], 2016, ARXIV161204251
  • [6] [Anonymous], 2019, CAT AD MPEG DASH
  • [7] [Anonymous], 2016, PROC INT C MACH LEAR
  • [8] [Anonymous], 2017, IEEE INT S BROADB MU, DOI DOI 10.1109/BMSB.2017.7986143
  • [9] [Anonymous], 2018, FDN TRENDS ROBOT, DOI DOI 10.1561/2300000053
  • [10] [Anonymous], 2018, REINFORCEMENT LEARNI