TY - GEN
T1 - MobiRNN
T2 - 1st International Workshop on Deep Learning for Mobile Systems and Applications, EMDL 2017, Co-located with MobiSys 2017
AU - Cao, Qingqing
AU - Balasubramanian, Niranjan
AU - Balasubramanian, Aruna
N1 - Publisher Copyright:
© 2017 ACM.
PY - 2017/6/23
Y1 - 2017/6/23
N2 - In this paper, we explore optimizations to run Recurrent Neural Network (RNN) models locally on mobile devices. RNN models are widely used for Natural Language Processing, Machine translation, and other tasks. However, existing mobile applications that use RNN models do so on the cloud. To address privacy and efficiency concerns, we show how RNN models can be run locally on mobile devices. Existing work on porting deep learning models to mobile devices focus on Convolution Neural Networks (CNNs) and cannot be applied directly to RNN models. In response, we present MobiRNN, a mobile-specific optimization framework that implements GPU offloading specifically for mobile GPUs. Evaluations using an RNN model for activity recognition shows that MobiRNN does significantly decrease the latency of running RNN models on phones.
AB - In this paper, we explore optimizations to run Recurrent Neural Network (RNN) models locally on mobile devices. RNN models are widely used for Natural Language Processing, Machine translation, and other tasks. However, existing mobile applications that use RNN models do so on the cloud. To address privacy and efficiency concerns, we show how RNN models can be run locally on mobile devices. Existing work on porting deep learning models to mobile devices focus on Convolution Neural Networks (CNNs) and cannot be applied directly to RNN models. In response, we present MobiRNN, a mobile-specific optimization framework that implements GPU offloading specifically for mobile GPUs. Evaluations using an RNN model for activity recognition shows that MobiRNN does significantly decrease the latency of running RNN models on phones.
UR - https://www.scopus.com/pages/publications/85029715180
U2 - 10.1145/3089801.3089804
DO - 10.1145/3089801.3089804
M3 - Conference contribution
AN - SCOPUS:85029715180
T3 - EMDL 2017 - Proceedings of the 1st International Workshop on Deep Learning for Mobile Systems and Applications, co-located with MobiSys 2017
SP - 1
EP - 6
BT - EMDL 2017 - Proceedings of the 1st International Workshop on Deep Learning for Mobile Systems and Applications, co-located with MobiSys 2017
PB - Association for Computing Machinery
Y2 - 23 June 2017 through 23 June 2017
ER -