This work presents CascadeCNN, an automated toolflow that pushes the quantisation limits of any given CNN model, to perform high-throughput inference by exploiting the computation time-accuracy trade-off. Without the need for retraining, a two-stage architecture tailored for any given FPGA device is generated, consisting of a low- and a high-precision unit. A confidence evaluation unit is employed between them to identify misclassified cases at run time and forward them to the high-precision unit or terminate computation. Experiments demonstrate that CascadeCNN achieves a performance boost of up to 55% for VGG-16 and 48% for AlexNet over the baseline design for the same resource budget and accuracy.
@article{arxiv.1805.08743,
title = {CascadeCNN: Pushing the performance limits of quantisation},
author = {Alexandros Kouris and Stylianos I. Venieris and Christos-Savvas Bouganis},
journal= {arXiv preprint arXiv:1805.08743},
year = {2018}
}