Otsuka, Jun and Saigo, Hayato (2022) The process theory of causality: an overview. [Preprint]

Preview

Text Process_theory_of_causatlity.pdf Download (1MB) | Preview

Abstract

This article offers an informal overview of the category-theoretical approach to causal modeling introduced by Jacobs et al. (2019) and explores some of its conceptual as well as methodological implications. The categorical formalism emphasizes the aspect of causality as a process, and represents a causal system as a network of connected mechanisms. We show that this alternative perspective sheds new light on the long-standing issue regarding the validity of the Markov condition, and also provides a formal mapping between micro-level causal models and abstracted macro models.

---

Export/Citation:

EndNote | BibTeX | Dublin Core | ASCII/Text Citation (Chicago) | HTML Citation | OpenURL

Social Networking:

Share |

---

Item Type:	Preprint
Creators:	Creators Email ORCID Otsuka, Jun junotk@gmail.com 0000-0003-4774-9740 Saigo, Hayato harmoniahayato@gmail.com
Keywords:	Symmetric monoidal category, String diagrams, Markov condition, Abstraction, Causal representation learning
Subjects:	General Issues > Causation Specific Sciences > Artificial Intelligence > Machine Learning
Depositing User:	Jun Otsuka
Date Deposited:	14 Oct 2022 13:23
Last Modified:	14 Oct 2022 13:23
Item ID:	21267
Subjects:	General Issues > Causation Specific Sciences > Artificial Intelligence > Machine Learning
Date:	October 2022
URI:	https://philsci-archive.pitt.edu/id/eprint/21267

Monthly Views for the past 3 years

Monthly Downloads for the past 3 years

Plum Analytics

Actions (login required)

View Item