IEEE Internet of Things Journal · 2024 · 11 citations · 15 references
ComputingComputer EngineeringComputer ScienceArtificial IntelligenceEmbedded SystemsMachine LearningDeep LearningHardware AccelerationKernel ArchitectureNeural NetworksOptimization For Machine LearningProcessor MicroarchitectureConvolutionConvolution KernelsConvolutional Neural NetworksDeep Feature LearningDeep InferenceDeep Learning AccelerationMicrokernelsOptimization Algorithms
We perform a series of optimisations on the convolution operator within the ARM CMSIS-NN library to improve the performance of deep learning tasks on Arduino development boards equipped with ARM Cortex-M4 and M7 microcontrollers. To this end, we develop custom microkernels that efficiently handle the internal computations required by the convolution operator via the lowering approach and the direct method, and we design two techniques to avoid register spilling. We also take advantage of all the RAM on the Arduino boards by reusing it as a scratchpad for the convolution filters. The integration of these techniques into CMSIS-NN, when invoked by TensorFlow Lite for microcontrollers for quantised versions of VGG, SqueezeNet, ResNet, and MobileNet-like convolutional neural networks enhances the overall inference speed by a factor ranging from 1.13× to 1.50×.
15
Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren et al. · 2016 · 222.1K citations · Full text
Efficient Processing of Deep Neural Networks: A Tutorial and Survey
Vivienne Sze, Yu‐Hsin Chen, Tien-Ju Yang et al. · Proceedings of the IEEE · 2017 · 4K citations