Poster: Optimizing Mobile Video Telephony Using Deep Imitation Learning.

Despite the pervasive use of real-time video telephony services, their quality of experience (QoE) remains unsatisfactory, especially over the mobile Internet. We conduct a largescale measurement campaign on Taobao-Live, an operational mobile video telephony service. Our analysis shows that the application-layer video codec and transport-layer protocols remain highly uncoordinated, which represents one major reason for the low QoE. We thus propose Concerto, a machine learning based framework to resolve the issue. We train Concerto with the massive data traces from the measurement campaign using a custom-designed imitation learning algorithm, which enables Concerto to learn from past experience following an expert's iterative demonstration/supervision. We have implemented and incorporated Concerto into the Taobao-Live. Our experiments show that Concerto outperforms state-of-the-art solutions, improving video quality while reducing stalling time by multi-folds under various practical scenarios. [ABSTRACT FROM AUTHOR]