Neural architecture search (NAS) has been very successful at outperforming
human-designed convolutional neural networks (CNN) in accuracy, and when
hardware information is present, latency as well. However, NAS-designed CNNs
typically have a complicated topology, therefore, it may be difficult to design
a custom hardware (HW) accelerator for such CNNs. We automate HW-CNN codesign
using NAS by including parameters from both the CNN model and the HW
accelerator, and we jointly search for the best model-accelerator pair that
boosts accuracy and efficiency. We call this Codesign-NAS. In this paper we
focus on defining the Codesign-NAS multiobjective optimization problem,
demonstrating its effectiveness, and exploring different ways of navigating the
codesign search space. For CIFAR-10 image classification, we enumerate close to
4 billion model-accelerator pairs, and find the Pareto frontier within that
large search space. This allows us to evaluate three different
reinforcement-learning-based search strategies. Finally, compared to ResNet on
its most optimal HW accelerator from within our HW design space, we improve on
CIFAR-100 classification accuracy by 1.3% while simultaneously increasing
performance/area by 41% in just~1000 GPU-hours of running Codesign-NAS.

本研究利用神经网络结构搜索技术，并联合考虑卷积神经网络模型和硬件加速器的参数，实现自动化的硬件 - 卷积神经网络代码设计。此方法能有效提高分类精度和效率，显著优于现有设计，本研究着重于在代码设计空间中进行多目标优化问题的定义以及探索使用强化学习的搜索策略，使得在 1000 个 GPU 小时内，相比于 ResNet 在优化硬件加速器情况下在 CIFAR-100 分类准确率上提高了 1.3％，同时提高了 41％的性能 / 面积比。