Distributional Computational Graphs: Error Bounds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Elias, Olof Hallqvist, Selby, Michael, Stanley-Marbell, Phillip
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912898455961600
author Elias, Olof Hallqvist
Selby, Michael
Stanley-Marbell, Phillip
author_facet Elias, Olof Hallqvist
Selby, Michael
Stanley-Marbell, Phillip
contents We study a general framework of distributional computational graphs: computational graphs whose inputs are probability distributions rather than point values. We analyze the discretization error that arises when these graphs are evaluated using finite approximations of continuous probability distributions. Such an approximation might be the result of representing a continuous real-valued distribution using a discrete representation or from constructing an empirical distribution from samples (or might be the output of another distributional computational graph). We establish non-asymptotic error bounds in terms of the Wasserstein-1 distance, without imposing structural assumptions on the computational graph.
format Preprint
id arxiv_https___arxiv_org_abs_2601_16250
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Distributional Computational Graphs: Error Bounds
Elias, Olof Hallqvist
Selby, Michael
Stanley-Marbell, Phillip
Machine Learning
Computational Engineering, Finance, and Science
Numerical Analysis
Probability
Primary 60H10, 65C30, Secondary 60J60, 65C05
We study a general framework of distributional computational graphs: computational graphs whose inputs are probability distributions rather than point values. We analyze the discretization error that arises when these graphs are evaluated using finite approximations of continuous probability distributions. Such an approximation might be the result of representing a continuous real-valued distribution using a discrete representation or from constructing an empirical distribution from samples (or might be the output of another distributional computational graph). We establish non-asymptotic error bounds in terms of the Wasserstein-1 distance, without imposing structural assumptions on the computational graph.
title Distributional Computational Graphs: Error Bounds
topic Machine Learning
Computational Engineering, Finance, and Science
Numerical Analysis
Probability
Primary 60H10, 65C30, Secondary 60J60, 65C05
url https://arxiv.org/abs/2601.16250