![[2602.15586] Uniform error bounds for quantized dynamical models](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.15586] Uniform error bounds for quantized dynamical models
Summary
This paper presents uniform error bounds for quantized dynamical models, providing statistical guarantees on their accuracy when learned from dependent data sequences, particularly in hybrid system identification.
Why It Matters
Understanding the accuracy of quantized dynamical models is crucial for applications in system identification, especially in contexts where hardware constraints impact performance. This research contributes to the field by offering interpretable statistical complexities that can guide practical implementations.
Key Takeaways
- Develops uniform error bounds for quantized models in dynamical systems.
- Introduces slow-rate and fast-rate bounds applicable to imperfect optimization algorithms.
- Translates hardware constraints into statistical complexities for better model understanding.
Computer Science > Machine Learning arXiv:2602.15586 (cs) [Submitted on 17 Feb 2026] Title:Uniform error bounds for quantized dynamical models Authors:Abdelkader Metakalard (CRAN, SYNALP), Fabien Lauer (SYNALP, LORIA), Kevin Colin (CRAN), Marion Gilson (CRAN) View a PDF of the paper titled Uniform error bounds for quantized dynamical models, by Abdelkader Metakalard (CRAN and 5 other authors View PDF Abstract:This paper provides statistical guarantees on the accuracy of dynamical models learned from dependent data sequences. Specifically, we develop uniform error bounds that apply to quantized models and imperfect optimization algorithms commonly used in practical contexts for system identification, and in particular hybrid system identification. Two families of bounds are obtained: slow-rate bounds via a block decomposition and fast-rate, variance-adaptive, bounds via a novel spaced-point strategy. The bounds scale with the number of bits required to encode the model and thus translate hardware constraints into interpretable statistical complexities. Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML) Cite as: arXiv:2602.15586 [cs.LG] (or arXiv:2602.15586v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2602.15586 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journal reference: IFAC Journal of Systems and Control, 2026, 35, pp.100373 Submission history From: Abdelkader Metakalard [view email] [via CCSD proxy] [v1] Tue, 1...
