English

Loss-Complexity Landscape and Model Structure Functions

Information Theory 2025-10-20 v4 Artificial Intelligence Machine Learning Mathematical Physics math.IT math.MP

Abstract

We develop a framework for dualizing the Kolmogorov structure function hx(α)h_x(\alpha), which then allows using computable complexity proxies. We establish a mathematical analogy between information-theoretic constructs and statistical mechanics, introducing a suitable partition function and free energy functional. We explicitly prove the Legendre-Fenchel duality between the structure function and free energy, showing detailed balance of the Metropolis kernel, and interpret acceptance probabilities as information-theoretic scattering amplitudes. A susceptibility-like variance of model complexity is shown to peak precisely at loss-complexity trade-offs interpreted as phase transitions. Practical experiments with linear and tree-based regression models verify these theoretical predictions, explicitly demonstrating the interplay between the model complexity, generalization, and overfitting threshold.

Keywords

Cite

@article{arxiv.2507.13543,
  title  = {Loss-Complexity Landscape and Model Structure Functions},
  author = {Alexander Kolpakov},
  journal= {arXiv preprint arXiv:2507.13543},
  year   = {2025}
}

Comments

25 pages, 11 figures; GitHub repository at https://github.com/sashakolpakov/structure-functions

R2 v1 2026-07-01T04:07:02.444Z