Minimum Width of Leaky-ReLU Neural Networks for Uniform Universal Approximation
This provides a theoretical foundation for neural network architecture design in approximating continuous functions, though it is incremental as it builds on prior work on critical widths for L^p functions.
The paper tackles the problem of determining the minimum width of leaky-ReLU neural networks for uniform universal approximation of continuous functions, finding that the exact minimum width is max(d_x, d_y) + Δ(d_x, d_y), where Δ accounts for additional dimensions needed for diffeomorphism-based approximations.
The study of universal approximation properties (UAP) for neural networks (NN) has a long history. When the network width is unlimited, only a single hidden layer is sufficient for UAP. In contrast, when the depth is unlimited, the width for UAP needs to be not less than the critical width $w^*_{\min}=\max(d_x,d_y)$, where $d_x$ and $d_y$ are the dimensions of the input and output, respectively. Recently, \cite{cai2022achieve} shows that a leaky-ReLU NN with this critical width can achieve UAP for $L^p$ functions on a compact domain ${K}$, \emph{i.e.,} the UAP for $L^p({K},\mathbb{R}^{d_y})$. This paper examines a uniform UAP for the function class $C({K},\mathbb{R}^{d_y})$ and gives the exact minimum width of the leaky-ReLU NN as $w_{\min}=\max(d_x,d_y)+Δ(d_x, d_y)$, where $Δ(d_x, d_y)$ is the additional dimensions for approximating continuous functions with diffeomorphisms via embedding. To obtain this result, we propose a novel lift-flow-discretization approach that shows that the uniform UAP has a deep connection with topological theory.