arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2309.13445cs.ARcs.AIeess.SP
收藏

AxOMaP:基于数学规划设计FPGA近似算术算子

AxOMaP: Designing FPGA-based Approximate Arithmetic Operators using Mathematical Programming

Interuniversity Microelectronics Centre (IMEC)(校际微电子中心);Center for Advancing Electronics Dresden (CfAED)(德累斯顿先进电子中心);Technische Universität Dresden(德累斯顿工业大学)

Siva Satyendra Sahoo, Salim Ullah, Akash Kumar

AI总结:

本文提出AxOMaP方法,通过数据分析驱动的混合整数二次约束规划引导进化优化搜索,在FPGA有符号8位乘法器设计中实现PPA与行为精度的联合优化,超体积指标较传统进化算法提升最高21%。

摘要:

随着机器学习(ML)算法在嵌入式系统中的应用日益广泛,为这些资源受限系统设计低成本计算机算术的需求不断上升。因此,利用此类算法固有容错特性的新兴计算模型,如近似计算和随机计算,正被积极探索用于在资源受限系统上实现ML推理。近似计算(AxC)旨在通过允许一定程度的行为精度(BEHAV)降低,为应用带来功耗、性能和面积(PPA)方面的超比例收益。使用近似算子(AxO)进行计算机算术是实现AxC较为普遍的方法之一。与仅对计算机算术进行精度缩放相比,AxO为更细粒度的优化提供了额外空间。为此,设计面向特定平台且具有成本效益的近似算子构成了一项重要的研究目标。近年来,多项工作报道了使用基于AI/ML的方法来综合新型的基于FPGA的AxO。然而,大多数此类工作将AI/ML的使用局限于设计迭代优化过程中使用的基于ML的代理函数。为此,我们提出了一种新颖的数据分析驱动的、基于数学规划的方法,用于为FPGA综合近似算子。具体而言,我们基于特征描述数据的相关性分析结果,构建混合整数二次约束规划,并利用其解为进化优化算法提供更具方向性的搜索方法。与传统的基于进化算法的优化相比,在有符号8位乘法器设计中联合优化PPA和BEHAV时,我们报告超体积指标提升高达21%。

英文摘要:

With the increasing application of machine learning (ML) algorithms in embedded systems, there is a rising necessity to design low-cost computer arithmetic for these resource-constrained systems. As a result, emerging models of computation, such as approximate and stochastic computing, that leverage the inherent error-resilience of such algorithms are being actively explored for implementing ML inference on resource-constrained systems. Approximate computing (AxC) aims to provide disproportionate gains in the power, performance, and area (PPA) of an application by allowing some level of reduction in its behavioral accuracy (BEHAV). Using approximate operators (AxOs) for computer arithmetic forms one of the more prevalent methods of implementing AxC. AxOs provide the additional scope for finer granularity of optimization, compared to only precision scaling of computer arithmetic. To this end, designing platform-specific and cost-efficient approximate operators forms an important research goal. Recently, multiple works have reported using AI/ML-based approaches for synthesizing novel FPGA-based AxOs. However, most of such works limit usage of AI/ML to designing ML-based surrogate functions used during iterative optimization processes. To this end, we propose a novel data analysis-driven mathematical programming-based approach to synthesizing approximate operators for FPGAs. Specifically, we formulate mixed integer quadratically constrained programs based on the results of correlation analysis of the characterization data and use the solutions to enable a more directed search approach for evolutionary optimization algorithms. Compared to traditional evolutionary algorithms-based optimization, we report up to 21% improvement in the hypervolume, for joint optimization of PPA and BEHAV, in the design of signed 8-bit multipliers.

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。

↑