Accelerating Convolutional Neural Networks via Activation Map Compression
This work addresses efficiency challenges for deploying neural networks in resource-constrained environments, representing an incremental improvement over existing methods.
The paper tackles the problem of high computational and memory requirements for convolutional neural networks on low-powered devices by proposing a three-stage compression and acceleration pipeline, achieving up to 1.6x acceleration and 6x compression with accuracy increases of up to 0.55% on datasets like ImageNet and CIFAR-10.
The deep learning revolution brought us an extensive array of neural network architectures that achieve state-of-the-art performance in a wide variety of Computer Vision tasks including among others, classification, detection and segmentation. In parallel, we have also been observing an unprecedented demand in computational and memory requirements, rendering the efficient use of neural networks in low-powered devices virtually unattainable. Towards this end, we propose a three-stage compression and acceleration pipeline that sparsifies, quantizes and entropy encodes activation maps of Convolutional Neural Networks. Sparsification increases the representational power of activation maps leading to both acceleration of inference and higher model accuracy. Inception-V3 and MobileNet-V1 can be accelerated by as much as $1.6\times$ with an increase in accuracy of $0.38\%$ and $0.54\%$ on the ImageNet and CIFAR-10 datasets respectively. Quantizing and entropy coding the sparser activation maps lead to higher compression over the baseline, reducing the memory cost of the network execution. Inception-V3 and MobileNet-V1 activation maps, quantized to $16$ bits, are compressed by as much as $6\times$ with an increase in accuracy of $0.36\%$ and $0.55\%$ respectively.