arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.12858cs.NIcs.AI
收藏

基于强化学习的真实世界视频自适应

Real-world Video Adaptation with Reinforcement Learning

MIT Computer Science and Artificial Intelligence Laboratory(麻省理工学院计算机科学与人工智能实验室);Facebook

Hongzi Mao, Shannon Chen, Drew Dimmery, Shaun Singh, Drew Blaisdell, Yuandong Tian, Mohammad Alizadeh, Eytan Bakshy

摘要:

客户端视频播放器采用自适应比特率(ABR)算法来优化用户体验质量(QoE)。我们评估最近提出的基于RL的ABR方法在Facebook的基于Web的视频流媒体平台。现实世界的ABR包含几个挑战,需要定制的设计超出了现成的RL算法-我们实现了一个可扩展的神经网络架构,支持视频与任意比特率编码;我们设计了一种训练方法,以应对由网络条件的随机性造成的差异;我们利用约束贝叶斯优化奖励塑造,以优化冲突的QoE目标。在为期一周的全球部署中,超过3000万个视频流会话,我们的RL方法优于现有的人工ABR算法。

英文摘要:

Client-side video players employ adaptive bitrate (ABR) algorithms to optimize user quality of experience (QoE). We evaluate recently proposed RL-based ABR methods in Facebook's web-based video streaming platform. Real-world ABR contains several challenges that requires customized designs beyond off-the-shelf RL algorithms -- we implement a scalable neural network architecture that supports videos with arbitrary bitrate encodings; we design a training method to cope with the variance resulting from the stochasticity in network conditions; and we leverage constrained Bayesian optimization for reward shaping in order to optimize the conflicting QoE objectives. In a week-long worldwide deployment with more than 30 million video streaming sessions, our RL approach outperforms the existing human-engineered ABR algorithms.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑