Robust Design Space Modeling

Robust Design Space Modeling
复制标题

稳健的设计空间建模

DOI:
10.1145/2668118
复制
发表时间:
2015-03
期刊:
ACM Trans. Design Autom. Electr. Syst.
影响因子:
--
通讯作者:
Yunji Chen
Yunji Chen
中科院分区:
其他
文献类型:
--
作者:
Olivier Temam;Ling Li;Depei Qian;Yunji Chen

文献摘要

参考文献

相似文献

微处理器的体系结构设计空间相对于待决的处理器参数通常是指数地大。为了避免模拟设计空间中的所有配置,已经利用机器学习和统计技术来构建回归模型,用于表征架构配置和响应之间的关系(例如,性能或功耗)。然而,这篇文章表明,许多学习技术在不同的设计空间和基准上的准确性差异可能足以误导决策。这清楚地表明,将在以前的建模任务(每个任务都涉及设计空间,基准和设计目标)上工作良好的技术应用于新任务的风险很高,因此强大的工具可能不切实际。受机器学习领域集成学习的启发,我们提出了一个强大的框架ELSE,以减少设计空间建模的准确性变化。而不是采用单一的学习技术,在以前的调查,ELSE采用不同的学习技术,为每个建模任务建立多个基础回归模型。这不是不同技术的简单组合(例如,总是相信具有最小误差的回归模型)。相反,ELSE谨慎地维护了基本回归模型的多样性,并从基本模型中构建了一个元模型,即使在基本模型远远不准确的情况下,该元模型也可以提供准确的预测。因此,我们能够减少最终预测误差大到不可接受的情况。实验结果验证了ELSE的鲁棒性:与广泛使用的人工神经网络相比,在52个不同的建模任务,ELSE减少了约62%的精度变化。此外,ELSE减少了27%和85%的平均预测误差的MIPS和POWER设计空间,分别调查。
Architectural design spaces of microprocessors are often exponentially large with respect to the pending processor parameters. To avoid simulating all configurations in the design space, machine learning and statistical techniques have been utilized to build regression models for characterizing the relationship between architectural configurations and responses (e.g., performance or power consumption). However, this article shows that the accuracy variability of many learning techniques over different design spaces and benchmarks can be significant enough to mislead the decision-making. This clearly indicates a high risk of applying techniques that work well on previous modeling tasks (each involving a design space, benchmark, and design objective) to a new task, due to which the powerful tools might be impractical. Inspired by ensemble learning in the machine learning domain, we propose a robust framework called ELSE to reduce the accuracy variability of design space modeling. Rather than employing a single learning technique as in previous investigations, ELSE employs distinct learning techniques to build multiple base regression models for each modeling task. This is not a trivial combination of different techniques (e.g., always trusting the regression model with the smallest error). Instead, ELSE carefully maintains the diversity of base regression models and constructs a metamodel from the base models that can provide accurate predictions even when the base models are far from accurate. Consequently, we are able to reduce the number of cases in which the final prediction errors are unacceptably large. Experimental results validate the robustness of ELSE: compared with the widely used artificial neural network over 52 distinct modeling tasks, ELSE reduces the accuracy variability by about 62%. Moreover, ELSE reduces the average prediction error by 27% and 85% for the investigated MIPS and POWER design spaces, respectively.
DOI: 10.4135/9781452244723.n136
发表时间: 1996
期刊: --
影响因子: --
作者:
P. Adriaans;Dolf Zantinge
通讯作者: P. Adriaans;Dolf Zantinge
DOI: 10.1109/ispass.2007.363742
发表时间: 2007-04
期刊: 2007 IEEE International Symposium on Performance Analysis of Systems & Software
影响因子: --
作者:
ElMoustapha Ould-Ahmed-Vall;J. Woodlee;Charles R. Yount;K. Doshi;S. Abraham
通讯作者: ElMoustapha Ould-Ahmed-Vall;J. Woodlee;Charles R. Yount;K. Doshi;S. Abraham
DOI: 10.1145/2514641.2514647
发表时间: 2013-09
期刊: ACM Trans. Embed. Comput. Syst.
影响因子: --
作者:
Giovanni Mariani;G. Palermo;V. Zaccaria;C. Silvano
通讯作者: Giovanni Mariani;G. Palermo;V. Zaccaria;C. Silvano
DOI: --
发表时间: 2014
期刊: Journal of management science
影响因子: --
作者:
อนิรุธ สืบสิงห์
通讯作者: อนิรุธ สืบสิงห์
DOI: 10.1016/s0893-6080(05)80023-1
发表时间: 1992-01-01
期刊: NEURAL NETWORKS
影响因子: 7.8
作者:
WOLPERT, DH
通讯作者: WOLPERT, DH