arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.05271cs.ARcs.AIcs.ETcs.LG
收藏

用于训练深度神经网络的混合存内计算架构

Hybrid In-memory Computing Architecture for the Training of Deep Neural Networks

King’s College London(伦敦国王学院);Arizona State University(亚利桑那州立大学)

Vinay Joshi, Wangxin He, Jae-sun Seo, Bipin Rajendran

摘要:

在冯-诺依曼架构上训练深度神经网络(DNN)所涉及的成本促使开发用于高效DNN训练加速器的新解决方案。我们提出了一种混合存内计算(HIC)架构,用于在硬件加速器上训练DNN,从而实现内存高效推理,并在基准任务中优于基线软件准确性。我们引入了一个权重表示技术,利用二进制和多级相变存储器(PCM)设备,这导致了一个内存高效的推理加速器。与以前的基于存内计算的实现不同,我们使用了一个低精度的权重更新累加器,从而节省了更多的内存。我们训练ResNet-32网络使用HIC对CIFAR-10图像进行分类。对于可比较的模型大小,通过利用适当的网络宽度乘数,基于FPGA的训练优于以浮点32位(FP 32)精度训练的基线网络。此外,我们观察到,基于神经网络的训练导致推理模型大小减少约50%,以实现基线可比的准确性。我们还表明,PCM设备中的时间漂移对长时间(年)的训练后推理准确性的影响可以忽略不计。最后,我们的模拟表明,基于FPGA的训练自然确保了设备所看到的写擦除周期数是PCM耐久性极限的一小部分,证明了这种架构实现可以在现场学习的硬件平台的可行性。

英文摘要:

The cost involved in training deep neural networks (DNNs) on von-Neumann architectures has motivated the development of novel solutions for efficient DNN training accelerators. We propose a hybrid in-memory computing (HIC) architecture for the training of DNNs on hardware accelerators that results in memory-efficient inference and outperforms baseline software accuracy in benchmark tasks. We introduce a weight representation technique that exploits both binary and multi-level phase-change memory (PCM) devices, and this leads to a memory-efficient inference accelerator. Unlike previous in-memory computing-based implementations, we use a low precision weight update accumulator that results in more memory savings. We trained the ResNet-32 network to classify CIFAR-10 images using HIC. For a comparable model size, HIC-based training outperforms baseline network, trained in floating-point 32-bit (FP32) precision, by leveraging appropriate network width multiplier. Furthermore, we observe that HIC-based training results in about 50% less inference model size to achieve baseline comparable accuracy. We also show that the temporal drift in PCM devices has a negligible effect on post-training inference accuracy for extended periods (year). Finally, our simulations indicate HIC-based training naturally ensures that the number of write-erase cycles seen by the devices is a small fraction of the endurance limit of PCM, demonstrating the feasibility of this architecture for achieving hardware platforms that can learn in the field.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑