English

Analog Physical Systems Can Exhibit Double Descent

Disordered Systems and Neural Networks 2025-11-25 v1 Machine Learning

Abstract

An important component of the success of large AI models is double descent, in which networks avoid overfitting as they grow relative to the amount of training data, instead improving their performance on unseen data. Here we demonstrate double descent in a decentralized analog network of self-adjusting resistive elements. This system trains itself and performs tasks without a digital processor, offering potential gains in energy efficiency and speed -- but must endure component non-idealities. We find that standard training fails to yield double descent, but a modified protocol that accommodates this inherent imperfection succeeds. Our findings show that analog physical systems, if appropriately trained, can exhibit behaviors underlying the success of digital AI. Further, they suggest that biological systems might similarly benefit from over-parameterization.

Keywords

Cite

@article{arxiv.2511.17825,
  title  = {Analog Physical Systems Can Exhibit Double Descent},
  author = {Sam Dillavou and Jason W Rocks and Jacob F Wycoff and Andrea J Liu and Douglas J Durian},
  journal= {arXiv preprint arXiv:2511.17825},
  year   = {2025}
}

Comments

11 pages 7 figures

R2 v1 2026-07-01T07:49:50.136Z