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 1712.07447cs.NEcs.PL
收藏

数据流矩阵机与V值:程序与神经网络之间的桥梁

Dataflow Matrix Machines and V-values: a Bridge between Programs and Neural Nets

Michael Bukatin, Jon Anthony

摘要:

1)数据流矩阵机(DMM)通过用线性流(支持线性组合的流)替换数字流来推广神经网络,允许激活函数的任意输入和输出,具有有限动态可变活动部分的可数大小网络能够无限增长,以及非常有表现力的自引用机制。2)DMM适用于通用编程,同时保留了递归神经网络的关键属性:程序通过实数矩阵表示,并且这些矩阵的连续变化会在相关程序中产生任意小的变化。3)V值空间(基于嵌套映射的类向量元素)特别有用,使DMM具有可变激活函数并方便地表示传统的数据结构。

英文摘要:

1) Dataflow matrix machines (DMMs) generalize neural nets by replacing streams of numbers with linear streams (streams supporting linear combinations), allowing arbitrary input and output arities for activation functions, countable-sized networks with finite dynamically changeable active part capable of unbounded growth, and a very expressive self-referential mechanism. 2) DMMs are suitable for general-purpose programming, while retaining the key property of recurrent neural networks: programs are expressed via matrices of real numbers, and continuous changes to those matrices produce arbitrarily small variations in the associated programs. 3) Spaces of V-values (vector-like elements based on nested maps) are particularly useful, enabling DMMs with variadic activation functions and conveniently representing conventional data structures.

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。

↑