The Laplace Microarchitecture for Tracking Data Uncertainty and Its Implementation in a RISC-V Processor

The Laplace Microarchitecture for Tracking Data Uncertainty and Its Implementation in a RISC-V Processor
复制标题

用于跟踪数据不确定性的拉普拉斯微架构及其在 RISC-V 处理器中的实现

DOI:
10.1145/3466752.3480131
复制
发表时间:
2021
期刊:
--
影响因子:
--
通讯作者:
Tsoutsouras V
Tsoutsouras V
中科院分区:
--
文献类型:
--
作者:
Tsoutsouras V

文献摘要

参考文献

被引文献

相似文献

我们提出了拉普拉斯,一个微体系结构,用于跟踪与体系结构状态配对的概率分布的机器表示。我们提出了两种新的处理器内分布表示方法,它们是概率分布的近似,就像浮点数表示是实数的近似一样。拉普拉斯执行未经修改的RISC-V二进制文件,并可以通过它们跟踪不确定性。我们提出了两组ISA扩展,以提供一种机制来初始化微体系结构中的分布信息,并允许应用程序在不暴露ISA上的不确定性表示的情况下查询分布信息的统计信息。我们使用一套21个基准测试来评估LaPlace的精度和性能,范围从变分量子算法、传感器数据处理到材料属性建模。在基准测试上的蒙特卡洛模拟平均需要2-076倍的指令(在某些情况下高达21-343倍)才能达到拉普拉斯在一次执行中所能达到的精度。与最先进的蒙特卡罗替代方案相比,拉普拉斯获得了比Pacal[22]平均1.3倍的精度改进,与NIST不确定性机器[26]使用的方法(使用到蒙特卡洛的沃瑟斯坦距离进行量化)相比,精度提高了4.6倍以上。与需要用特定于领域的语言重写软件或进行广泛的源代码级别更改的现有不确定性跟踪方法不同,拉普拉斯实现了所有这些优势,同时不需要更改现有的二进制文件来跟踪它们的不确定性,只需极小的更改即可将不确定性信息输入微体系结构。我们已经部署了LaPlace的实现,将其作为可通过云访问的虚拟机形式的商业产品。
We present Laplace, a microarchitecture for tracking machine representations of probability distributions paired with architectural state. We present two new methods for in-processor distribution representations which are approximations of probability distributions just as floating-point number representations are approximations of real-valued numbers. Laplace executes unmodified RISC-V binaries and can track uncertainty through them. We present two sets of ISA extensions to provide a mechanism to initialize distributional information in the microarchitecture and to allow applications to query statistics of the distributional information without exposing the uncertainty representations above the ISA.We evaluate the accuracy and performance of Laplace using a suite of 21 benchmarks spanning domains ranging from variational quantum algorithms and sensor data processing to materials properties modeling. Monte Carlo simulation on the benchmarks requires 2 076 × more instructions on average (and up to 21 343 × in some cases) to achieve the same accuracy that Laplace can achieve in a single execution. Compared to state-of-the-art alternatives to Monte Carlo, Laplace achieves an average 1.3 × accuracy improvement versus PaCAL [22] and more than 4.6 × accuracy improvement versus the method used by the NIST Uncertainty Machine [26], quantified using the Wasserstein distance to Monte Carlo.Unlike existing methods for uncertainty tracking which require software to be rewritten in a domain-specific language or extensive source-level changes, Laplace achieves all of these benefits while requiring no changes to existing binaries in order to track uncertainty through them, with only minimal changes required to get uncertainty information into the microarchitecture. We have deployed an implementation of Laplace as a commercial product in the form of a cloud-accessible virtual machine.
使用智能手机传感器进行现实生活人类活动识别的公共领域数据集
DOI: --
发表时间: 2020
期刊: Italian National Conference on Sensors
影响因子: --
作者:
Daniel Garcia;D. Rivero;Enrique Fernández;M. R. Luaces
通讯作者: M. R. Luaces
DOI: 10.1145/2813885.2737959
发表时间: 2015
期刊: Sigplan Notices
影响因子: --
作者:
P. Panchekha;Alex Sanchez;James R. Wilcox;Zachary Tatlock
通讯作者: Zachary Tatlock
RISC 处理器中模糊指令的评估
DOI: --
发表时间: 1993
期刊: [Proceedings 1993] Second IEEE International Conference on Fuzzy Systems
影响因子: --
作者:
Hiroyuki Watanabe;David Chen
通讯作者: David Chen
使用不确定数据进行编程的抽象和技术
DOI: --
发表时间: 2013
期刊:
影响因子: --
作者:
James Bornholt
通讯作者: James Bornholt
模糊处理器架构设计的 RISC 方法
DOI: --
发表时间: 1992
期刊: [1992 Proceedings] IEEE International Conference on Fuzzy Systems
影响因子: --
作者:
Hiroyuki Watanabe
通讯作者: Hiroyuki Watanabe