Towards representation agnostic probabilistic programming

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fenske, Ole, Popko, Maximilian, Bader, Sebastian, Kirste, Thomas
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914224652943360
author Fenske, Ole
Popko, Maximilian
Bader, Sebastian
Kirste, Thomas
author_facet Fenske, Ole
Popko, Maximilian
Bader, Sebastian
Kirste, Thomas
contents Current probabilistic programming languages and tools tightly couple model representations with specific inference algorithms, preventing experimentation with novel representations or mixed discrete-continuous models. We introduce a factor abstraction with five fundamental operations that serve as a universal interface for manipulating factors regardless of their underlying representation. This enables representation-agnostic probabilistic programming where users can freely mix different representations (e.g. discrete tables, Gaussians distributions, sample-based approaches) within a single unified framework, allowing practical inference in complex hybrid models that current toolkits cannot adequately express.
format Preprint
id arxiv_https___arxiv_org_abs_2512_23740
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards representation agnostic probabilistic programming
Fenske, Ole
Popko, Maximilian
Bader, Sebastian
Kirste, Thomas
Programming Languages
Artificial Intelligence
Current probabilistic programming languages and tools tightly couple model representations with specific inference algorithms, preventing experimentation with novel representations or mixed discrete-continuous models. We introduce a factor abstraction with five fundamental operations that serve as a universal interface for manipulating factors regardless of their underlying representation. This enables representation-agnostic probabilistic programming where users can freely mix different representations (e.g. discrete tables, Gaussians distributions, sample-based approaches) within a single unified framework, allowing practical inference in complex hybrid models that current toolkits cannot adequately express.
title Towards representation agnostic probabilistic programming
topic Programming Languages
Artificial Intelligence
url https://arxiv.org/abs/2512.23740