Major commercial client-side video players employ adaptive bitrate (ABR) algorithms to improve user quality of experience (QoE). With the evolvement of ABR algorithms, increasingly complex methods such as neural networks have been adopted to pursue better performance. However, these complex methods are too heavyweight to be directly implemented in client devices, especially mobile phones with very limited resources. Existing solutions suffer from a tradeoff between algorithm performance and deployment overhead. To make the implementation of sophisticated ABR algorithms practical, we propose PiTree, a general, high-performance and scalable framework that can faithfully convert sophisticated ABR algorithms into lightweight decision trees to reduce deployment overhead. We also provide a theoretical analysis on the upper bound of performance loss during conversion. Evaluation results on three representative ABR algorithms demonstrate that PiTree could faithfully convert ABR algorithms into decision trees with less than 3% average performance loss. Moreover, comparing to original implementation solutions, PiTree could save operating expenses by up to millions of dollars every month for large content providers.


PiTree: Practical Implementation of ABR Algorithms Using Decision Trees
Zili Meng, Jing Chen, Yaning Guo, Chen Sun, Hongxin Hu, Mingwei Xu
In Proceedings of the ACM International Conference on Multimedia (2019) [Paper][ACM DL]

Practically Deploying Heavyweight Adaptive Bitrate Algorithms With Teacher-Student Learning
Zili Meng, Yaning Guo, Yixin Shen, Jing Chen, Chao Zhou, Minhu Wang, Jia Zhang, Mingwei Xu, Chen Sun, Hongxin Hu
In IEEE/ACM Transactions on Networking [Paper][IEEEXplore]


 author = {Meng, Zili and Chen, Jing and Guo, Yaning and Sun, Chen and Hu, Hongxin and Xu, Mingwei},
 title = {PiTree: Practical Implementation of ABR Algorithms Using Decision Trees},
 year = {2019},
 url = {https://doi.org/10.1145/3343031.3350866},
 booktitle = {Proceedings of the 27th ACM International Conference on Multimedia},
 pages = {2431–2439},
 location = {Nice, France},
 series = {MM ’19}

  title={Practically Deploying Heavyweight Adaptive Bitrate Algorithms With Teacher-Student Learning},
  author={Meng, Zili and Guo, Yaning and Shen, Yixin and Chen, Jing and Zhou, Chao and Wang, Minhu and Zhang, Jia and Xu, Mingwei and Sun, Chen and Hu, Hongxin},
  journal={IEEE/ACM Transactions on Networking},






This project is supported by National Key R&D Program of China (2017YFB0801701) and National Science Foundation of China (61625203, 61832013, 61872426).


For any questions, please send an email to zilim@ieee.org.