Skip to yearly menu bar Skip to main content


TREC: Transient Redundancy Elimination-based Convolution

Jiawei Guan · Feng Zhang · Jiesong Liu · Hsin-Hsuan Sung · Ruofan Wu · Xiaoyong Du · Xipeng Shen

Hall J (level 1) #910

Keywords: [ convolution acceleration ] [ Transient redundancy ] [ back-propagation ]


The intensive computations in convolutional neural networks (CNNs) pose challenges for resource-constrained devices; eliminating redundant computations from convolution is essential. This paper gives a principled method to detect and avoid transient redundancy, a type of redundancy existing in input data or activation maps and hence changing across inferences. By introducing a new form of convolution (TREC), this new method makes transient redundancy detection and avoidance an inherent part of the CNN architecture, and the determination of the best configurations for redundancy elimination part of CNN backward propagation. We provide a rigorous proof of the robustness and convergence of TREC-equipped CNNs. TREC removes over 96% computations and achieves 3.51x average speedups on microcontrollers with minimal (about 0.7%) accuracy loss.

Chat is not available.