Freeborn, David Peter Wallis (2026) A Model of Understanding in Deep Learning Systems. [Preprint]

Text Deep_Learning_World_Models_Structural_Understanding.pdf Download (1MB)

Abstract

I propose a model of systematic understanding, suitable for machine learning systems. On this account, an agent understands a property of a target system when it contains an adequate internal model that tracks real regularities, is coupled to the target by stable bridge principles, and supports reliable prediction. I argue that contemporary deep learning systems often can and do achieve such understanding. However they generally fall short of the ideal of scientific understanding: the understanding is symbolically misaligned with the target system, not explicitly reductive, and only weakly unifying. I label this the Fractured Understanding Hypothesis.

---

Export/Citation:

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

Social Networking:

Share |

---

Item Type:	Preprint
Creators:	Creators Email ORCID Freeborn, David Peter Wallis david.freeborn@nulondon.ac.uk 0000-0002-2117-8145
Keywords:	deep learning; machine learning; artificial intelligence; philosophy of artificial intelligence; philosophy of science; scientific understanding; understanding; representation; memorization and generalization; interpolation and extrapolation; mechanistic interpretability; world models; philosophy of artificial intelligence; neural networks; memorization; generalization; mechanistic interpretability; unification
Subjects:	Specific Sciences > Computation/Information Specific Sciences > Computer Science Specific Sciences > Artificial Intelligence Specific Sciences > Cognitive Science > Concepts and Representations General Issues > Explanation Specific Sciences > Cognitive Science > Learning and Memory Specific Sciences > Artificial Intelligence > Machine Learning General Issues > Models and Idealization
Depositing User:	David Freeborn
Date Deposited:	06 Apr 2026 12:25
Last Modified:	06 Apr 2026 12:25
Item ID:	28921
Subjects:	Specific Sciences > Computation/Information Specific Sciences > Computer Science Specific Sciences > Artificial Intelligence Specific Sciences > Cognitive Science > Concepts and Representations General Issues > Explanation Specific Sciences > Cognitive Science > Learning and Memory Specific Sciences > Artificial Intelligence > Machine Learning General Issues > Models and Idealization
Date:	5 April 2026
URI:	https://philsci-archive.pitt.edu/id/eprint/28921

Monthly Views for the past 3 years

Monthly Downloads for the past 3 years

Plum Analytics

Actions (login required)

View Item