A Novel Schur-Decomposition-Based Weight Projection Method for Stable State-Space Neural-Network Architectures

• URL: http://arxiv.org/abs/2605.14489v1
• Date: Thu, 14 May 2026 07:28:11 GMT
• Title: A Novel Schur-Decomposition-Based Weight Projection Method for Stable State-Space Neural-Network Architectures
• Abstract summary: We introduce a novel stability-ensuring and backpropagation-compatible projection scheme based on the Schur decomposition.<n>The proposed methods dynamically project the quasi-triangular factor of the state matrix's real Schur decomposition onto its nearest stable peer.<n>Experiments on synthetic linear systems demonstrate that the method achieves accuracy and convergence rates comparable to those of state-of-the-art stable-system identification techniques.
• Score: 1.1151457846264183
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Building black-box models for dynamical systems from data is a challenging problem in machine learning, especially when asymptotic stability guarantees are required. In this paper, we introduce a novel stability-ensuring and backpropagation-compatible projection scheme based on the Schur decomposition for the state matrix of linear discrete-time state-space layers, as well as an alternative pre-factorized formulation of the methodology. The proposed methods dynamically project the quasi-triangular factor of the state matrix's real Schur decomposition onto its nearest stable peer, ensuring stable dynamics with minimal overparameterization. Experiments on synthetic linear systems demonstrate that the method achieves accuracy and convergence rates comparable to those of state-of-the-art stable-system identification techniques, despite a marginal increase in computational complexity. Furthermore, the lower weight count facilitates convergence during training without sacrificing accuracy in stacked neural-network architectures with static nonlinearities targeting real-world datasets. These results suggest that the Schur-based projection provides a numerically robust framework for identifying complex dynamics on par with the State of the Art while satisfying strict asymptotic-stability requirements.

Related papers

• Learning Spatio-Temporal Foundation Models from Pure Synthetic Data [85.23980441738811]
  NeoST is a scalable synthetic-training corpus to real-world bias, a latent-space reasoning architecture that generates and iteratively refines multiple future trajectories, and latent-space objectives that emphasize structural dynamics and enable inference-time correction under distribution shifts.<n>Extensive experiments across diverse real-world benchmarks show that NeoST consistently outperforms existing STFMs in diverse real-worldtemporal systems.
  arXiv  Detail & Related papers  (2026-06-27T02:56:46Z)
• A Quadratic Order Reduction -- Gaussian Process Ordinary Differential Equation framework for the inference of Large Continuous Dynamical Systems [0.0]
  We propose a framework for forecasting complex dynamical systems via a kernel autonomous ordinary differential equation approach.<n>We integrate it with quadratic order reduced-order modelling and sphere projection for learning the latent dynamics efficiently while preserving stability.<n>Results demonstrate the potential of the framework as a robust and stable tool for forecasting complex dynamical systems with rigorous uncertainty quantification.
  arXiv  Detail & Related papers  (2026-06-11T08:49:18Z)
• Robust Learning of Heterogeneous Dynamic Systems [3.5271214584237427]
  We propose a novel distributionally robust learning approach for modeling heterogeneous ODE systems.<n>We show the resulting estimator admits an explicit weighted average representation, where the weights are obtained from a quadratic optimization.<n>We establish rigorous theoretical guarantees for the proposed method, including consistency of the stabilized weights, error bound for robust trajectory estimation, and robustness validity of pointwise confidence interval.
  arXiv  Detail & Related papers  (2026-04-07T00:40:55Z)
• ODELoRA: Training Low-Rank Adaptation by Solving Ordinary Differential Equations [54.886931928255564]
  Low-rank adaptation (LoRA) has emerged as a widely adopted parameter-efficient fine-tuning method in deep transfer learning.<n>We propose a novel continuous-time optimization dynamic for LoRA factor matrices in the form of an ordinary differential equation (ODE)<n>We show that ODELoRA achieves stable feature learning, a property that is crucial for training deep neural networks at different scales of problem dimensionality.
  arXiv  Detail & Related papers  (2026-02-07T10:19:36Z)
• Interpretable Neural Approximation of Stochastic Reaction Dynamics with Guaranteed Reliability [4.736119820998459]
  We introduce DeepSKA, a neural framework that achieves interpretability, guaranteed reliability, and substantial computational gains.<n>DeepSKA yields mathematically transparent representations that generalise across states, times, and output functions, and it integrates this structure with a small number of simulations to produce unbiased, provably convergent, and dramatically lower-magnitude estimates than classical Monte Carlo.
  arXiv  Detail & Related papers  (2025-12-06T04:45:31Z)
• Stable-by-Design Neural Network-Based LPV State-Space Models for System Identification [6.5745172279769255]
  We propose a neural network-based state-space model that simultaneously learns latent states and internal scheduling variables.<n>The state-transition matrix is guaranteed to be stable through a Schur-based parameterization.<n>The proposed NN-SS is evaluated on benchmark nonlinear systems, and the results demonstrate that the model consistently matches or surpasses classical subspace identification methods.
  arXiv  Detail & Related papers  (2025-10-21T10:25:54Z)
• Randomized Space-Time Sampling for Affine Graph Dynamical Systems [3.8379346559198235]
  We consider signals evolving over time according to a linear dynamical system on a graph, where both the initial state and the source term are bandlimited.<n>We introduce two random space-time sampling regimes and analyze the conditions under which stable recovery is achievable.
  arXiv  Detail & Related papers  (2025-09-20T21:43:05Z)
• Efficient Transformed Gaussian Process State-Space Models for Non-Stationary High-Dimensional Dynamical Systems [49.819436680336786]
  We propose an efficient transformed Gaussian process state-space model (ETGPSSM) for scalable and flexible modeling of high-dimensional, non-stationary dynamical systems.<n>Specifically, our ETGPSSM integrates a single shared GP with input-dependent normalizing flows, yielding an expressive implicit process prior that captures complex, non-stationary transition dynamics.<n>Our ETGPSSM outperforms existing GPSSMs and neural network-based SSMs in terms of computational efficiency and accuracy.
  arXiv  Detail & Related papers  (2025-03-24T03:19:45Z)
• Stability properties of gradient flow dynamics for the symmetric low-rank matrix factorization problem [22.648448759446907]
  We show that a low-rank factorization serves as a building block in many learning tasks. We offer new insight into the shape of the trajectories associated with local search parts of the dynamics.
  arXiv  Detail & Related papers  (2024-11-24T20:05:10Z)
• Distributionally Robust Model-based Reinforcement Learning with Large State Spaces [55.14361269378122]
  Three major challenges in reinforcement learning are the complex dynamical systems with large state spaces, the costly data acquisition processes, and the deviation of real-world dynamics from the training environment deployment. We study distributionally robust Markov decision processes with continuous state spaces under the widely used Kullback-Leibler, chi-square, and total variation uncertainty sets. We propose a model-based approach that utilizes Gaussian Processes and the maximum variance reduction algorithm to efficiently learn multi-output nominal transition dynamics.
  arXiv  Detail & Related papers  (2023-09-05T13:42:11Z)
• Numerically Stable Sparse Gaussian Processes via Minimum Separation using Cover Trees [57.67528738886731]
  We study the numerical stability of scalable sparse approximations based on inducing points. For low-dimensional tasks such as geospatial modeling, we propose an automated method for computing inducing points satisfying these conditions.
  arXiv  Detail & Related papers  (2022-10-14T15:20:17Z)
• A Priori Denoising Strategies for Sparse Identification of Nonlinear Dynamical Systems: A Comparative Study [68.8204255655161]
  We investigate and compare the performance of several local and global smoothing techniques to a priori denoise the state measurements. We show that, in general, global methods, which use the entire measurement data set, outperform local methods, which employ a neighboring data subset around a local point.
  arXiv  Detail & Related papers  (2022-01-29T23:31:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.