TY - GEN
T1 - Play and Prune
T2 - 28th International Joint Conference on Artificial Intelligence, IJCAI 2019
AU - Singh, Pravendra
AU - Verma, Vinay Kumar
AU - Rai, Piyush
AU - Namboodiri, Vinay P.
PY - 2019/1/1
Y1 - 2019/1/1
N2 - While convolutional neural networks (CNN) have achieved impressive performance on various classification/recognition tasks, they typically consist of a massive number of parameters. This results in significant memory requirement as well as computational overheads. Consequently, there is a growing need for filter-level pruning approaches for compressing CNN based models that not only reduce the total number of parameters but reduce the overall computation as well. We present a new min-max framework for filter-level pruning of CNNs. Our framework, called Play and Prune (PP), jointly prunes and fine-tunes CNN model parameters, with an adaptive pruning rate, while maintaining the model's predictive performance. Our framework consists of two modules: (1) An adaptive filter pruning (AFP) module, which minimizes the number of filters in the model; and (2) A pruning rate controller (PRC) module, which maximizes the accuracy during pruning. Moreover, unlike most previous approaches, our approach allows directly specifying the desired error tolerance instead of pruning level. Our compressed models can be deployed at run-time, without requiring any special libraries or hardware. Our approach reduces the number of parameters of VGG-16 by an impressive factor of 17.5X, and number of FLOPS by 6.43X, with no loss of accuracy, significantly outperforming other state-of-the-art filter pruning methods.
AB - While convolutional neural networks (CNN) have achieved impressive performance on various classification/recognition tasks, they typically consist of a massive number of parameters. This results in significant memory requirement as well as computational overheads. Consequently, there is a growing need for filter-level pruning approaches for compressing CNN based models that not only reduce the total number of parameters but reduce the overall computation as well. We present a new min-max framework for filter-level pruning of CNNs. Our framework, called Play and Prune (PP), jointly prunes and fine-tunes CNN model parameters, with an adaptive pruning rate, while maintaining the model's predictive performance. Our framework consists of two modules: (1) An adaptive filter pruning (AFP) module, which minimizes the number of filters in the model; and (2) A pruning rate controller (PRC) module, which maximizes the accuracy during pruning. Moreover, unlike most previous approaches, our approach allows directly specifying the desired error tolerance instead of pruning level. Our compressed models can be deployed at run-time, without requiring any special libraries or hardware. Our approach reduces the number of parameters of VGG-16 by an impressive factor of 17.5X, and number of FLOPS by 6.43X, with no loss of accuracy, significantly outperforming other state-of-the-art filter pruning methods.
UR - http://www.scopus.com/inward/record.url?scp=85074905499&partnerID=8YFLogxK
M3 - Chapter in a published conference proceeding
AN - SCOPUS:85074905499
T3 - IJCAI International Joint Conference on Artificial Intelligence
SP - 3460
EP - 3466
BT - Proceedings of the 28th International Joint Conference on Artificial Intelligence, IJCAI 2019
A2 - Kraus, Sarit
PB - International Joint Conferences on Artificial Intelligence
Y2 - 10 August 2019 through 16 August 2019
ER -