The realization function of a shallow ReLU network is a continuous and piecewise affine function $f:\mathbb R^d\to \mathbb R$, where the domain $\mathbb R^{d}$ is partitioned by a set of $n$ hyperplanes into cells on which $f$ is affine. We show that the minimal representation for $f$ uses either $n$, $n+1$ or $n+2$ neurons and we characterize each of the three cases. In the particular case, where the input layer is one-dimensional, minimal representations always use at most $n+1$ neurons but in all higher dimensional settings there are functions for which $n+2$ neurons are needed. Then we show that the set of minimal networks representing $f$ forms a $C^\infty$-submanifold $M$ and we derive the dimension and the number of connected components of $M$. Additionally, we give a criterion for the hyperplanes that guarantees that all continuous, piecewise affine functions are realization functions of appropriate ReLU networks.
翻译:浅 ReLU 网络的实现功能是一个连续的和零星的折叠函数 $f:\mathbb R ⁇ d\ to\ mathbb R$, 域名$\mathbb R ⁇ d} 被一组一美元高空飞机分割成以美元为折合金的单元格。 我们显示, 美元的最低代表值是美元, 美元+1美元或美元+2美元的神经元, 我们对这三种情况都有特征。 在特定情况下, 输入层是一维的, 最小代表值总是在最多为n+1美元的神经元使用, 但在所有较高维环境中, 都存在需要n+2美元的神经元的函数。 然后, 我们显示, 代表美元为 美元的最小网络组构成 $C ⁇ inty- submany $M$, 我们从中得出维度和连接的元元数。 此外, 我们给出一个超标准, 用于超标准, 保证所有连续的、 直方形的连接功能是适当的RELU 网络的实现功能。