RAELLA:为高效、低分辨率、低损耗模拟PIM改革算术:无需重新训练!
RAELLA: Reforming the Arithmetic for Efficient, Low-Resolution, and Low-Loss Analog PIM: No Retraining Required!
AI总结:
RAELLA架构通过权重编码、自适应切片和输入动态切片降低模拟值分辨率,从而在无需重新训练DNN的情况下使用低分辨率ADC,实现高效低损耗模拟PIM推理。
摘要:
存内处理(PIM)加速器通过减少昂贵的数据移动并使用阻变RAM(ReRAM)进行高效的模拟计算,有潜力高效运行深度神经网络(DNN)推理。遗憾的是,PIM加速器的整体效率受到能耗密集的模数转换器(ADC)的限制。此外,现有降低ADC成本的加速器要么通过改变DNN权重,要么通过使用降低输出保真度的低分辨率ADC来实现。这些策略会损害DNN精度和/或需要昂贵的DNN重新训练来补偿。为了解决这些问题,我们提出了RAELLA架构。RAELLA使架构适应每个DNN;它通过编码权重以产生接近零的模拟值来降低所计算模拟值的分辨率,为每个DNN层自适应地切片权重,并通过推测与恢复动态切片输入。低分辨率模拟值使RAELLA既能使用高效的低分辨率ADC,又能在无需重新训练的情况下保持精度,同时以更少的ADC转换次数进行计算。与其他低精度损耗PIM加速器相比,RAELLA将能效提高了最多4.9倍,吞吐量提高了最多3.3倍。与导致精度损失并重新训练DNN以恢复精度的PIM加速器相比,RAELLA在无需昂贵的DNN重新训练的情况下实现了类似的效率和吞吐量。
英文摘要:
Processing-In-Memory (PIM) accelerators have the potential to efficiently run Deep Neural Network (DNN) inference by reducing costly data movement and by using resistive RAM (ReRAM) for efficient analog compute. Unfortunately, overall PIM accelerator efficiency is limited by energy-intensive analog-to-digital converters (ADCs). Furthermore, existing accelerators that reduce ADC cost do so by changing DNN weights or by using low-resolution ADCs that reduce output fidelity. These strategies harm DNN accuracy and/or require costly DNN retraining to compensate. To address these issues, we propose the RAELLA architecture. RAELLA adapts the architecture to each DNN; it lowers the resolution of computed analog values by encoding weights to produce near-zero analog values, adaptively slicing weights for each DNN layer, and dynamically slicing inputs through speculation and recovery. Low-resolution analog values allow RAELLA to both use efficient low-resolution ADCs and maintain accuracy without retraining, all while computing with fewer ADC converts. Compared to other low-accuracy-loss PIM accelerators, RAELLA increases energy efficiency by up to 4.9$\times$ and throughput by up to 3.3$\times$. Compared to PIM accelerators that cause accuracy loss and retrain DNNs to recover, RAELLA achieves similar efficiency and throughput without expensive DNN retraining.
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。