Real-world Video Adaptation with Reinforcement Learning

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

Apr 30, 2019 ICML 2019 Workshop RL4RealLife Submission readers: everyone
  • Abstract: 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 prior ABR algorithm.
0 Replies

Loading