The Information-Theoretic Imperative: Compression and the Epistemic Foundations of Intelligence

Christian Dittrich, Jennifer Flygare Kinne · 2025

Read The Information-Theoretic Imperative: Compression and the Epistemic Foundations of Intelligence by Christian Dittrich, Jennifer Flygare Kinne (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-25883.txt.

Description

Christian Dittrich and Jennifer Flygare Kinne argue that compression is a core force behind how both brains and deep networks form their inner representations. A representation is the pattern a system uses to encode what it senses. Task trained networks predict primate visual brain responses well, even though the systems differ in material and training, and the authors seek a deeper reason than shared images or tasks alone.

Their Compression Efficiency Principle says that codes built on unstable shortcuts pay a growing cost when conditions shift, while codes built on stable features spread that cost out. The view ties brain energy limits and sensory coding to deep learning scale, shortcut failures, and data augmentation, and it yields testable predictions.

Catalog record

CREATORChristian Dittrich, Jennifer Flygare Kinne
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE107 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-25883

Browse the full library · About the Tower of Babel