A Simple Method to Reduce Off-chip Memory Accesses on Convolutional Neural Networks
arXiv:1901.09614
Abstract
For convolutional neural networks, a simple algorithm to reduce off-chip memory accesses is proposed by maximally utilizing on-chip memory in a neural process unit. Especially, the algorithm provides an effective way to process a module which consists of multiple branches and a merge layer. For Inception-V3 on Samsung's NPU in Exynos, our evaluation shows that the proposed algorithm makes off-chip memory accesses reduced by 1/50, and accordingly achieves 97.59 % reduction in the amount of feature-map data to be transferred from/to off-chip memory.
9 pages, 10 figures, under review (by ICML2019)
References in corpus (10)
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- MobileNetV2: Inverted Residuals and Linear Bottlenecks
- BinaryConnect: Training Deep Neural Networks with binary weights during propagations
- Deep Learning with Limited Numerical Precision
- Fixed Point Quantization of Deep Convolutional Networks
- Hardware-oriented Approximation of Convolutional Neural Networks
- ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
- DSD: Dense-Sparse-Dense Training for Deep Neural Networks
- Reduced-Precision Strategies for Bounded Memory in Deep Neural Nets
- Convolutional Neural Network Quantization using Generalized Gamma Distribution