用可重构晶体管设计低功耗神经网络加速器，显著降低面积、延迟和能耗。
An Energy-Efficient RFET-Based Stochastic Computing Neural Network Accelerator
- 基于可重构晶体管设计高效随机计算单元，减少硬件开销
- 相比传统鳍式晶体管，在同工艺下面积、延迟、能耗均大幅降低
- 适合对能效要求高的边缘AI场景，如智能传感器
随机计算（SC）为传统卷积神经网络（CNN）提供了显著的硬件简化优势。然而，尽管具有这些优点，随机计算神经网络（SCNN）常因随机数生成器（SNGs）和累加并行计数器（APCs）等组件导致资源消耗过高，从而限制整体性能。本文提出一种基于可重构场效应晶体管（RFETs）的新型SCNN加速器。器件层面的可重构性使得SNGs、APCs及其他关键组件能够实现高度高效与紧凑的设计。为评估其系统级影响，采用现有代表性SCNN架构作为评测框架。基于公开的标准单元库，实验结果表明，所提出的基于RFET的SCNN加速器在相同技术节点下，相较于其鳍式晶体管（FinFET）对应方案，实现了显著的面积、延迟和能量消耗降低。
原文摘要 · Abstract (English)
Stochastic computing (SC) offers significant reductions in hardware complexity for traditional convolutional neural networks (CNNs). However, despite its advantages, stochastic computing neural networks (SCNNs) often suffer from high resource consumption due to components such as stochastic number generators (SNGs) and accumulative parallel counters (APCs), which limit overall performance. This paper proposes a novel SCNN accelerator based on reconfigurable field-effect transistors (RFETs). The inherent reconfigurability at the device level enables the design of highly efficient and compact SNGs, APCs, and other related essential components. To assess their system-level impact, a representative existing SCNN architecture is adopted as an evaluation framework. Based on accessible open-source standard cell libraries, experimental results demonstrate that the proposed RFET-based SCNN accelerator achieves significant reductions in area, latency, and energy consumption compared to its FinFET-based counterpart at the same technology node.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。