TY - GEN
T1 - Neural adaptive video streaming with pensieve
AU - Mao, Hongzi
AU - Netravali, Ravi
AU - Alizadeh, Mohammad
N1 - Funding Information:
Acknowledgments. We thank our shepherd, John Byers, and the anonymous SIGCOMM reviewers for their valuable feedback. We also thank Te-Yuan Huang for her guidance regarding video streaming in practice, and Jiaming Luo for fruitful discussions regarding the learning aspects of the design. This work was funded in part by NSF grants CNS-1617702, CNS-1563826, and CNS-1407470, the MIT Center for Wireless Networks and Mobile Computing, and a Qualcomm Innovation Fellowship.
Publisher Copyright:
© 2017 ACM.
PY - 2017/8/7
Y1 - 2017/8/7
N2 - Client-side video players employ adaptive bitrate (ABR) algorithms to optimize user quality of experience (QoE). Despite the abundance of recently proposed schemes, state-of-the-art ABR algorithms suffer from a key limitation: they use fixed control rules based on simplified or inaccurate models of the deployment environment. As a result, existing schemes inevitably fail to achieve optimal performance across a broad set of network conditions and QoE objectives. We propose Pensieve, a system that generates ABR algorithms using reinforcement learning (RL). Pensieve trains a neural network model that selects bitrates for future video chunks based on observations collected by client video players. Pensieve does not rely on pre-programmed models or assumptions about the environment. Instead, it learns to make ABR decisions solely through observations of the resulting performance of past decisions. As a result, Pensieve automatically learns ABR algorithms that adapt to a wide range of environments and QoE metrics.We compare Pensieve to state-of-theart ABR algorithms using trace-driven and real world experiments spanning a wide variety of network conditions, QoE metrics, and video properties. In all considered scenarios, Pensieve outperforms the best state-of-the-art scheme, with improvements in average QoE of 12%-25%. Pensieve also generalizes well, outperforming existing schemes even on networks for which it was not explicitly trained.
AB - Client-side video players employ adaptive bitrate (ABR) algorithms to optimize user quality of experience (QoE). Despite the abundance of recently proposed schemes, state-of-the-art ABR algorithms suffer from a key limitation: they use fixed control rules based on simplified or inaccurate models of the deployment environment. As a result, existing schemes inevitably fail to achieve optimal performance across a broad set of network conditions and QoE objectives. We propose Pensieve, a system that generates ABR algorithms using reinforcement learning (RL). Pensieve trains a neural network model that selects bitrates for future video chunks based on observations collected by client video players. Pensieve does not rely on pre-programmed models or assumptions about the environment. Instead, it learns to make ABR decisions solely through observations of the resulting performance of past decisions. As a result, Pensieve automatically learns ABR algorithms that adapt to a wide range of environments and QoE metrics.We compare Pensieve to state-of-theart ABR algorithms using trace-driven and real world experiments spanning a wide variety of network conditions, QoE metrics, and video properties. In all considered scenarios, Pensieve outperforms the best state-of-the-art scheme, with improvements in average QoE of 12%-25%. Pensieve also generalizes well, outperforming existing schemes even on networks for which it was not explicitly trained.
KW - Bitrate adaptation
KW - Reinforcement learning
KW - Video streaming
UR - http://www.scopus.com/inward/record.url?scp=85029435567&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85029435567&partnerID=8YFLogxK
U2 - 10.1145/3098822.3098843
DO - 10.1145/3098822.3098843
M3 - Conference contribution
AN - SCOPUS:85029435567
T3 - SIGCOMM 2017 - Proceedings of the 2017 Conference of the ACM Special Interest Group on Data Communication
SP - 197
EP - 210
BT - SIGCOMM 2017 - Proceedings of the 2017 Conference of the ACM Special Interest Group on Data Communication
PB - Association for Computing Machinery, Inc
T2 - 2017 Conference of the ACM Special Interest Group on Data Communication, SIGCOMM 2017
Y2 - 21 August 2017 through 25 August 2017
ER -