ReaLPrune:ReRAM Crossbar感知彩票修剪CNN
ReaLPrune: ReRAM Crossbar-aware Lottery Ticket Pruned CNNs
摘要:
在边缘训练机器学习(ML)模型(在最终用户设备上进行片上训练)可以解决许多紧迫的挑战,包括数据隐私/安全性,通过减少对通信结构和云基础设施的依赖来增加ML应用程序对世界不同地区的可访问性,并满足AR/VR应用程序的实时要求。然而,现有的边缘平台没有足够的计算能力来支持复杂的机器学习任务,例如训练大型CNN。基于ReRAM的架构为片上CNN训练/推理提供了高性能且节能的计算平台。然而,基于ReRAM的架构不能随着CNN的大小而扩展。较大的CNN具有更多的权重,这需要更多的ReRAM单元,而这些单元无法集成在单个芯片中。此外,在芯片上训练较大的CNN将需要更高的功率,这是这些较小的设备无法承受的。修剪是解决这一问题的有效方法。然而,现有的修剪技术要么只针对推理,或者它们不是交叉感知的。这导致基于ReRAM的架构上的CNN训练的次优硬件节省和性能优势。在本文中,我们通过提出一种新的横杆感知修剪策略来解决这个问题,称为ReaLPrune,它可以修剪超过90%的CNN权重。修剪后的模型可以从头开始训练,而不会有任何准确性损失。实验结果表明,与未修剪的CNN相比,ReaLPrune将硬件需求降低了77.2%,并将CNN训练速度加快了约20倍。ReaLPrune在性能和硬件节省方面也优于其他纵横感知修剪技术。此外,ReaLPrune对于多样化的数据集和更复杂的CNN同样有效
英文摘要:
Training machine learning (ML) models at the edge (on-chip training on end user devices) can address many pressing challenges including data privacy/security, increase the accessibility of ML applications to different parts of the world by reducing the dependence on the communication fabric and the cloud infrastructure, and meet the real-time requirements of AR/VR applications. However, existing edge platforms do not have sufficient computing capabilities to support complex ML tasks such as training large CNNs. ReRAM-based architectures offer high-performance yet energy efficient computing platforms for on-chip CNN training/inferencing. However, ReRAM-based architectures are not scalable with the size of the CNN. Larger CNNs have more weights, which requires more ReRAM cells that cannot be integrated in a single chip. Moreover, training larger CNNs on-chip will require higher power, which cannot be afforded by these smaller devices. Pruning is an effective way to solve this problem. However, existing pruning techniques are either targeted for inferencing only, or they are not crossbar-aware. This leads to sub-optimal hardware savings and performance benefits for CNN training on ReRAM-based architectures. In this paper, we address this problem by proposing a novel crossbar-aware pruning strategy, referred as ReaLPrune, which can prune more than 90% of CNN weights. The pruned model can be trained from scratch without any accuracy loss. Experimental results indicate that ReaLPrune reduces hardware requirements by 77.2% and accelerates CNN training by ~20X compared to unpruned CNNs. ReaLPrune also outperforms other crossbar-aware pruning techniques in terms of both performance and hardware savings. In addition, ReaLPrune is equally effective for diverse datasets and more complex CNNs
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。