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

Researchers propose the Compression Efficiency Principle (CEP) to explain why artificial neural networks and biological brains develop similar representations despite different substrates. The theory suggests both systems converge on efficient compression strategies that encode stable invariants rather than unstable correlations, providing a unified framework for understanding intelligence across biological and artificial systems.

• →The Compression Efficiency Principle explains convergence between AI and biological neural representations through shared compression strategies.
• →Representations that exploit unstable correlations pay an 'exception tax' while stable invariant-encoding representations are more efficient.
• →The framework connects biological metabolic constraints to AI phenomena like distribution shift failures and data augmentation benefits.
• →Theory predicts systematic coupling between compression efficiency and out-of-distribution robustness across different substrates.
• →Convergence between AI and biological intelligence reflects substrate-independent optimization basins rather than coincidence.