The Flow has not summarised this story yet — read it at arXiv Machine Learning.
The paper presents an in-context learning framework for system identification that replaces traditional two-step model estimation with a meta model mapping training data and query inputs directly to outputs. It introduces a probabilistic Transformer-based architecture capable of handling long training sequences and producing calibrated predictions over extended horizons. Additionally, the authors demonstrate that a pre-trained meta model can generate uncertainty-aware synthetic data to aid classical identification methods and validate their approach on standard nonlinear benchmarks and challenging out-of-distribution scenarios, achieving competitive accuracy and strong generalization.
arXiv:2505. 15497v3 Announce Type: replace Abstract: Neural networks hold great potential to act as approximate models of nonlinear dynamical systems, with the resulting neural approximations enabling verification and control of such systems.
The paper studies least squares parameter estimation for discrete‑time, unstable, closed‑loop nonlinear stochastic systems with linearly parametrised uncertainty and additive i.i.d. process noise. By perturbing the control policy with exploratory input and assuming a sub‑exponential input‑to‑state growth property, the authors derive non‑asymptotic bounds on the estimation error whenever the state trajectory remains in an informative region of the state space. When the entire state space is informative, the bounds hold with high probability for all time steps, and the authors illustrate the applicability of their results with examples that extend beyond existing work.
arXiv:2602. 12756v2 Announce Type: replace Abstract: Large Language Models (LLMs) have recently shown exceptional potential in time series forecasting (TSF), leveraging their inherent sequential reasoning capabilities to model complex temporal dynamics.
The paper discusses how offline-trained neural networks, when used in closed-loop control systems, can change the state distribution they were trained on, leading to performance degradation. It introduces a taxonomy of failure modes—Closed-Loop Identification Error, deployment-induced Action Error, and continuous Tracking Error—and uses discrete-time sensitivity analysis to explain how these errors arise. The authors propose time-scale separated online adaptation policies to mitigate distribution shift, validated on a nonlinear Duffing oscillator, showing that bounded continuous learning can prevent trajectory divergence while static deployments cannot.
arXiv:2506. 01226v3 Announce Type: replace-cross Abstract: We study parameterizations of stabilizing nonlinear policies for learning-based control.