Universal Regular Conditional Distributions

TL;DR

This paper introduces a deep learning model called the probabilistic transformer (PT) that can universally approximate regular conditional distributions, effectively handling high-dimensional data and avoiding the curse of dimensionality.

Contribution

The paper presents a novel neural network architecture that approximates any continuous function from Euclidean space to Wasserstein space, with strategies to mitigate the curse of dimensionality.

Findings

PT can approximate any continuous function from $ ^d$ to $ P_1( ^D)$ uniformly on compact sets.

The model avoids the curse of dimensionality through two different approximation strategies.

The approach extends attention mechanisms to probabilistic settings for universal approximation.

Abstract

We introduce a deep learning model that can universally approximate regular conditional distributions (RCDs). The proposed model operates in three phases: first, it linearizes inputs from a given metric space $\mathcal{X}$ to $\mathbb{R}^d$ via a feature map, then a deep feedforward neural network processes these linearized features, and then the network's outputs are then transformed to the $1$-Wasserstein space $\mathcal{P}_1(\mathbb{R}^D)$ via a probabilistic extension of the attention mechanism of Bahdanau et al.\ (2014). Our model, called the \textit{probabilistic transformer (PT)}, can approximate any continuous function from $\mathbb{R}^d $ to $\mathcal{P}_1(\mathbb{R}^D)$ uniformly on compact sets, quantitatively. We identify two ways in which the PT avoids the curse of dimensionality when approximating $\mathcal{P}_1(\mathbb{R}^D)$-valued functions. The first strategy builds functions in $C(\mathbb{R}^d,\mathcal{P}_1(\mathbb{R}^D))$ which can be efficiently approximated by a PT, uniformly on any given compact subset of $\mathbb{R}^d$. In the second approach, given any function $f$ in $C(\mathbb{R}^d,\mathcal{P}_1(\mathbb{R}^D))$, we build compact subsets of $\mathbb{R}^d$ whereon $f$ can be efficiently approximated by a PT.