Approximation of functions with one-bit neural networks

The celebrated universal approximation theorems for neural networks roughly state that any reasonable function can be arbitrarily well-approximated by a network whose parameters are appropriately chosen real numbers. This paper examines the approximation capabilities of one-bit neural networks -- those whose nonzero parameters are $\pm a$ for some fixed $a\not=0$. One of our main theorems shows that for any $f\in C^s([0,1]d)$ with $|f|\infty<1$ and error $\varepsilon$, there is a $f{NN}$ such that $|f(\boldsymbol{x})-f{NN}(\boldsymbol{x})|\leq \varepsilon$ for all $\boldsymbol{x}$ away from the boundary of $[0,1]^d$, and $f{NN}$ is either implementable by a ${\pm 1}$ quadratic network with $O(\varepsilon^{-2d/s})$ parameters or a ${\pm \frac 1 2 }$ ReLU network with $O(\varepsilon^{-2d/s}\log (1/\varepsilon))$ parameters, as $\varepsilon\to0$. We establish new approximation results for iterated multivariate Bernstein operators, error estimates for noise-shaping quantization on the Bernstein basis, and novel implementation of the Bernstein polynomials by one-bit quadratic and ReLU neural networks.