1: \begin{abstract}

2: Motivated by the developing mathematics of deep learning, we build universal functions approximators of continuous maps between arbitrary Polish metric spaces $\xxx$ and $\yyy$ using elementary functions between Euclidean spaces as building blocks. Earlier results assume that the target space $\yyy$ is a topological vector space. We overcome this limitation by ``randomization'': our approximators output discrete probability measures over $\yyy$. When $\xxx$ and $\yyy$ are Polish without additional structure, we prove very general qualitative guarantees; when they have suitable combinatorial structure, we prove quantitative guarantees for H\"{o}lder-like maps, including maps between finite graphs, solution operators to rough differential equations between certain Carnot groups, and continuous non-linear operators between Banach spaces arising in inverse problems. In particular, we show that the required number of Dirac measures is determined by the combinatorial structure of $\xxx$ and $\yyy$. For barycentric $\yyy$, including Banach spaces, $\mathbb{R}$-trees, Hadamard manifolds, or Wasserstein spaces on Polish metric spaces, our approximators reduce to $\yyy$-valued functions. When the Euclidean approximators are neural networks, our constructions generalize transformer networks, providing a new probabilistic viewpoint of geometric deep learning.

3: \end{abstract}

4: