A Short Information-Theoretic Analysis of Linear Auto-Regressive Learning

Abstract

In this note, we give a short information-theoretic proof of the consistency of the Gaussian maximum likelihood estimator in linear auto-regressive models. Our proof yields nearly optimal non-asymptotic rates for parameter recovery and works without any invocation of stability in the case of finite hypothesis classes.

Cite

Text

Ziemann. "A Short Information-Theoretic Analysis of Linear Auto-Regressive Learning." Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, 2025.

Markdown

[Ziemann. "A Short Information-Theoretic Analysis of Linear Auto-Regressive Learning." Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, 2025.](https://mlanthology.org/l4dc/2025/ziemann2025l4dc-short/)

BibTeX

@inproceedings{ziemann2025l4dc-short,
  title     = {{A Short Information-Theoretic Analysis of Linear Auto-Regressive Learning}},
  author    = {Ziemann, Ingvar},
  booktitle = {Proceedings of the 7th Annual Learning for Dynamics \& Control Conference},
  year      = {2025},
  pages     = {26-30},
  volume    = {283},
  url       = {https://mlanthology.org/l4dc/2025/ziemann2025l4dc-short/}
}