adaptive bitrate algorithms (ABR)

Practically Deploying Heavyweight Adaptive Bitrate Algorithms With Teacher-Student Learning
In TON 2020. The full journal version of PiTree, a general, high-performance, and scalable framework that can faithfully convert sophisticated ABR algorithms into decision trees with teacher-student learning. The journal version is completed with theoretical analysis on prediction error and real-world experiments.
Zili Meng, Yaning Guo, Yixin Shen, Jing Chen, Chao Zhou, Minhu Wang, Jia Zhang, Mingwei Xu, Chen Sun, Hongxin Hu
PDF Cite DOI
Practically Deploying Heavyweight Adaptive Bitrate Algorithms With Teacher-Student Learning
PiTree: Practical Implementation of ABR Algorithms Using Decision Trees
In Proc. ACM MM 2019. PiTree is a general, high-performance, and scalable framework to bridge the gap between offline design and online implementation of ABR algorithms using light-weight decision trees.
Zili Meng, Jing Chen, Yaning Guo, Chen Sun, Hongxin Hu, Mingwei Xu
PDF Cite Code Dataset Poster Slides DOI Website
PiTree: Practical Implementation of ABR Algorithms Using Decision Trees