arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.14652cs.DScs.DCcs.NE
收藏

用于矩阵乘法的恒深恒宽阈值电路

Constant-Depth and Subcubic-Size Threshold Circuits for Matrix Multiplication

Ojas Parekh, Cynthia A. Phillips, Conrad D. James, James B. Aimone

摘要:

McCulloch-Pitts阈值门布尔电路是20世纪后期作为一般计算模型而被大量研究的神经计算的经典模型。大规模神经计算硬件的最新进展使其实际实现成为近期的可能性。我们描述了一种理论方法,用于乘以两个$N$的$N$矩阵,集成了阈值门逻辑与传统的快速矩阵乘法算法,执行$O(N^ω)$算术运算的正常数$ω< 3$。我们的方法将这样一个快速矩阵乘法算法转换成一个恒定深度的阈值电路,大约$O(N^ω)$门。在此之前,我们的工作,这是不知道是否$Θ(N^3)$门矩阵乘法障碍是克服了恒定深度的阈值电路。稠密矩阵乘法是卷积神经网络训练中的核心运算。在神经架构上执行这项工作,而不是将其卸载到GPU上可能是一个有吸引力的选择。

英文摘要:

Boolean circuits of McCulloch-Pitts threshold gates are a classic model of neural computation studied heavily in the late 20th century as a model of general computation. Recent advances in large-scale neural computing hardware has made their practical implementation a near-term possibility. We describe a theoretical approach for multiplying two $N$ by $N$ matrices that integrates threshold gate logic with conventional fast matrix multiplication algorithms, that perform $O(N^ω)$ arithmetic operations for a positive constant $ω< 3$. Our approach converts such a fast matrix multiplication algorithm into a constant-depth threshold circuit with approximately $O(N^ω)$ gates. Prior to our work, it was not known whether the $Θ(N^3)$-gate barrier for matrix multiplication was surmountable by constant-depth threshold circuits. Dense matrix multiplication is a core operation in convolutional neural network training. Performing this work on a neural architecture instead of off-loading it to a GPU may be an appealing option.

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。

↑