Real-world Video Adaptation with Reinforcement Learning
This work addresses video streaming optimization for users on web-based platforms, representing an incremental improvement with domain-specific impact.
The paper tackled the problem of optimizing video streaming quality of experience (QoE) by evaluating and deploying a reinforcement learning (RL)-based adaptive bitrate (ABR) algorithm in Facebook's real-world platform, resulting in outperformance over existing human-engineered methods in a week-long deployment with over 30 million sessions.
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.