Uniform error bounds for quantized dynamical models

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.

• Publication year

  2026

• Venue

  IFAC Journal of Systems and Control

• Publication date

  2026-02-17

• Fields of study

  Mathematics, Computer Science, Engineering

• Identifiers
  DOI 10.1016/j.ifacsc.2026.100373arXiv 2602.15586
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• PAC-Bayes Generalisation Bounds for Dynamical Systems Including Stable RNNs

  2023cited by this paper
• Finite-Time Identification of Linear Systems: Fundamental Limits and Optimal Algorithms

  2023cited by this paper
• Learning Mixtures of Linear Dynamical Systems

  2022cited by this paper
• Statistical Learning Theory for Control: A Finite-Sample Perspective

  2022cited by this paper
• Learning with little mixing

  2022cited by this paper
• A statistical learning perspective on switched linear system identification

  2022influential reference
• Identification and Adaptive Control of Markov Jump Systems: Sample Complexity and Regret Bounds

  2021cited by this paper
• Statistical Learning Theory

  2021cited by this paper
• Learning Without Mixing: Towards A Sharp Analysis of Linear System Identification

  2018cited by this paper
• Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference

  2017cited by this paper
• Finite Time Identification in Unstable Linear Systems

  2017cited by this paper
• Deep Learning

  2016cited by this paper
• Estimating beta-mixing coefficients via histograms

  2011cited by this paper
• Estimating beta-mixing coefficients

  2011cited by this paper
• Model selection for weakly dependent time series forecasting

  2009cited by this paper
• Rademacher Complexity Bounds for Non-I.I.D. Processes

  2008cited by this paper
• A LEARNING THEORY APPROACH TO SYSTEM IDENTIFICATION

  2004cited by this paper
• Nonparametric Time Series Prediction Through Adaptive Model Selection

  2000cited by this paper
• Finite sample properties of system identification of ARX models under mixing conditions

  1999cited by this paper
• RATES OF CONVERGENCE FOR EMPIRICAL PROCESSES OF STATIONARY MIXING SEQUENCES

  1994influential reference
• Mixing: Properties and Examples

  1994cited by this paper

• No citing papers are available for this paper.