A Time-Series Foundation Model by Universal Delay Embedding

• URL: http://arxiv.org/abs/2509.12080v1
• Date: Mon, 15 Sep 2025 16:11:49 GMT
• Title: A Time-Series Foundation Model by Universal Delay Embedding
• Authors: Zijian Wang, Peng Tao, Jifan Shi, Rui Bao, Rui Liu, Luonan Chen,
• Abstract summary: This study introduces Universal Delay Embedding (UDE), a pretrained foundation model designed to revolutionize time-series forecasting.<n>UDE as a dynamical representation of observed data constructs two-dimensional subspace patches from Hankel matrices.<n>In particular, the learned dynamical representations and Koopman operator prediction forms from the patches exhibit exceptional interpretability.
• Score: 4.221753069966852
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This study introduces Universal Delay Embedding (UDE), a pretrained foundation model designed to revolutionize time-series forecasting through principled integration of delay embedding representation and Koopman operator prediction. Leveraging Takens' embedding theorem, UDE as a dynamical representation of observed data constructs two-dimensional subspace patches from Hankel matrices, theoretically preserving dynamical and topological properties of underlying dynamical systems. Such patches are viewed as images, which can be efficiently processed by exploiting advanced deep learning technologies. Computationally, these patches further serve as tokens for learning a self-attention encoder, thus enabling accurate prediction of nonlinear time-series by a finite-dimensional Koopman operator in a linear manner in a latent space. Extensive evaluations across various benchmarks and real-world climate datasets demonstrate over 20% average reduction in mean squared error versus state-of-the-art foundation models, alongside superior generalization in fine-tuning scenarios. In particular, the learned dynamical representations and Koopman operator prediction forms from the patches exhibit exceptional interpretability, with consistent identification of topologically informative subspaces and robust encoding of domain-invariant dynamics, establishing UDE as a scalable, interpretable framework for universal time-series modeling and forecasting with broad scientific and industrial applicability.

Related papers

• StepVAR: Structure-Texture Guided Pruning for Visual Autoregressive Models [98.72926158261937]
  We propose a training-free token pruning framework for Visual AutoRegressive models.<n>We employ a lightweight high-pass filter to capture local texture details, while leveraging Principal Component Analysis (PCA) to preserve global structural information.<n>To maintain valid next-scale prediction under sparse tokens, we introduce a nearest neighbor feature propagation strategy.
  arXiv  Detail & Related papers  (2026-03-02T11:35:05Z)
• EIDOS: Latent-Space Predictive Learning for Time Series Foundation Models [37.917978019436674]
  EIDOS is a foundation model family that shifts pretraining from future value prediction to latent-space predictive learning.<n>We train a causal Transformer to predict the evolution of latent representations, encouraging the emergence of structured and temporally coherent latent states.
  arXiv  Detail & Related papers  (2026-02-15T07:07:20Z)
• VFMF: World Modeling by Forecasting Vision Foundation Model Features [67.09340259579761]
  We introduce a generative forecaster that performs autoregressive flow matching in vision foundation models feature space.<n>We show that this latent information more effectively than previously used PCA-based alternatives, both for forecasting and other applications.<n>With matched architecture and compute, our method produces sharper and more accurate predictions than regression across all modalities.
  arXiv  Detail & Related papers  (2025-12-12T02:10:05Z)
• Learning Time-Aware Causal Representation for Model Generalization in Evolving Domains [50.66049136093248]
  We develop a time-aware structural causal model (SCM) that incorporates dynamic causal factors and the causal mechanism drifts.<n>We show that our method can yield the optimal causal predictor for each time domain.<n>Results on both synthetic and real-world datasets exhibit that SYNC can achieve superior temporal generalization performance.
  arXiv  Detail & Related papers  (2025-06-21T14:05:37Z)
• Spatial-Temporal-Spectral Unified Modeling for Remote Sensing Dense Prediction [20.1863553357121]
  Current deep learning architectures for remote sensing are fundamentally rigid.<n>We introduce the Spatial-Temporal-Spectral Unified Network (STSUN) for unified modeling.<n> STSUN can adapt to input and output data with arbitrary spatial sizes, temporal lengths, and spectral bands.<n>It unifies various dense prediction tasks and diverse semantic class predictions.
  arXiv  Detail & Related papers  (2025-05-18T07:39:17Z)
• Dynamical system prediction from sparse observations using deep neural networks with Voronoi tessellation and physics constraint [12.638698799995815]
  We introduce the Dynamic System Prediction from Sparse Observations using Voronoi Tessellation (DSOVT) framework. By integrating Voronoi tessellations with deep learning models, DSOVT is adept at predicting dynamical systems with sparse, unstructured observations. Compared to purely data-driven models, our physics-based approach enables the model to learn physical laws within explicitly formulated dynamics.
  arXiv  Detail & Related papers  (2024-08-31T13:43:52Z)
• Temporally Consistent Koopman Autoencoders for Forecasting Dynamical Systems [38.36312939874359]
  We introduce the temporally consistent Koopman autoencoder (tcKAE)<n>tcKAE generates accurate long-term predictions even with limited and noisy training data.<n>We empirically demonstrate tcKAE's superior performance over state-of-the-art KAE models across a variety of test cases.
  arXiv  Detail & Related papers  (2024-03-19T00:48:25Z)
• Generative Modeling of Regular and Irregular Time Series Data via Koopman VAEs [50.25683648762602]
  We introduce Koopman VAE, a new generative framework that is based on a novel design for the model prior. Inspired by Koopman theory, we represent the latent conditional prior dynamics using a linear map. KoVAE outperforms state-of-the-art GAN and VAE methods across several challenging synthetic and real-world time series generation benchmarks.
  arXiv  Detail & Related papers  (2023-10-04T07:14:43Z)
• Anamnesic Neural Differential Equations with Orthogonal Polynomial Projections [6.345523830122166]
  We propose PolyODE, a formulation that enforces long-range memory and preserves a global representation of the underlying dynamical system. Our construction is backed by favourable theoretical guarantees and we demonstrate that it outperforms previous works in the reconstruction of past and future data.
  arXiv  Detail & Related papers  (2023-03-03T10:49:09Z)
• Multivariate Time Series Forecasting with Dynamic Graph Neural ODEs [65.18780403244178]
  We propose a continuous model to forecast Multivariate Time series with dynamic Graph neural Ordinary Differential Equations (MTGODE) Specifically, we first abstract multivariate time series into dynamic graphs with time-evolving node features and unknown graph structures. Then, we design and solve a neural ODE to complement missing graph topologies and unify both spatial and temporal message passing.
  arXiv  Detail & Related papers  (2022-02-17T02:17:31Z)
• Differentiable Generalised Predictive Coding [2.868176771215219]
  This paper deals with differentiable dynamical models congruent with neural process theories that cast brain function as the hierarchical refinement of an internal generative model explaining observations. Our work extends existing implementations of gradient-based predictive coding and allows to integrate deep neural networks for non-linear state parameterization.
  arXiv  Detail & Related papers  (2021-12-02T22:02:56Z)
• Stochastically forced ensemble dynamic mode decomposition for forecasting and analysis of near-periodic systems [65.44033635330604]
  We introduce a novel load forecasting method in which observed dynamics are modeled as a forced linear system. We show that its use of intrinsic linear dynamics offers a number of desirable properties in terms of interpretability and parsimony. Results are presented for a test case using load data from an electrical grid.
  arXiv  Detail & Related papers  (2020-10-08T20:25:52Z)
• Target-Embedding Autoencoders for Supervised Representation Learning [111.07204912245841]
  This paper analyzes a framework for improving generalization in a purely supervised setting, where the target space is high-dimensional. We motivate and formalize the general framework of target-embedding autoencoders (TEA) for supervised prediction, learning intermediate latent representations jointly optimized to be both predictable from features as well as predictive of targets.
  arXiv  Detail & Related papers  (2020-01-23T02:37:10Z)

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.