A Model of Understanding in Deep Learning Systems

2026-04-05

AI Summary

This study addresses the challenge of defining and evaluating whether deep learning systems genuinely understand the properties of target systems. To this end, the authors develop a formal framework integrating philosophical epistemology with machine learning theory, proposing a “systematic understanding” model: understanding is achieved when an agent possesses an internal model that tracks genuine regularities, is stably coupled to the target system via bridging principles, and supports reliable predictions. Building on this framework, the paper advances the “fragmented understanding hypothesis,” arguing that while current deep learning systems exhibit limited forms of understanding, they fall significantly short of scientific understanding in terms of symbolic alignment, explicit reducibility, and unificatory power—thereby revealing a fundamental gap between artificial and scientific comprehension.

Technology Category

Machine Learning: Deep Learning TheoryCognitive Modeling & Cognitive Systems: Conceptual Inference and ReasoningPhilosophy and Ethics of AI: AI & Epistemology

Application Category

Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactionsSystems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applicationsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

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.

Problem

• understanding
• deep learning
• scientific understanding
• Fractured Understanding Hypothesis
• internal model

Innovation

• systematic understanding
• deep learning
• internal model
• bridge principles
• Fractured Understanding Hypothesis