Effective Dataset Distillation for Spatio-Temporal Forecasting with Bi-dimensional Compression

• URL: http://arxiv.org/abs/2603.10410v1
• Date: Wed, 11 Mar 2026 04:48:12 GMT
• Title: Effective Dataset Distillation for Spatio-Temporal Forecasting with Bi-dimensional Compression
• Abstract summary: We propose STemDist, the first dataset distillation method for S-temporal time series forecasting.<n>A key idea of our solution is to compress both temporal and spatial dimensions in a balanced manner, reducing training time and memory.
• Score: 26.189594254326334
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Spatio-temporal time series are widely used in real-world applications, including traffic prediction and weather forecasting. They are sequences of observations over extensive periods and multiple locations, naturally represented as multidimensional data. Forecasting is a central task in spatio-temporal analysis, and numerous deep learning methods have been developed to address it. However, as dataset sizes and model complexities continue to grow in practice, training deep learning models has become increasingly time- and resource-intensive. A promising solution to this challenge is dataset distillation, which synthesizes compact datasets that can effectively replace the original data for model training. Although successful in various domains, including time series analysis, existing dataset distillation methods compress only one dimension, making them less suitable for spatio-temporal datasets, where both spatial and temporal dimensions jointly contribute to the large data volume. To address this limitation, we propose STemDist, the first dataset distillation method specialized for spatio-temporal time series forecasting. A key idea of our solution is to compress both temporal and spatial dimensions in a balanced manner, reducing training time and memory. We further reduce the distillation cost by performing distillation at the cluster level rather than the individual location level, and we complement this coarse-grained approach with a subset-based granular distillation technique that enhances forecasting performance. On five real-world datasets, we show empirically that, compared to both general and time-series dataset distillation methods, datasets distilled by our STemDist method enable model training (1) faster (up to 6X) (2) more memory-efficient (up to 8X), and (3) more effective (with up to 12% lower prediction error).

Related papers

• SAS: Semantic-aware Sampling for Generative Dataset Distillation [55.27114962330541]
  We introduce a semantic-aware perspective for dataset distillation by leveraging Contrastive Language-Image Pretraining (CLIP) as a semantic prior for post-sampling.<n>Our goal is to obtain distilled datasets that are not only compact but also semantically class-discriminative and diverse.
  arXiv  Detail & Related papers  (2026-05-18T08:05:46Z)
• DIET: Learning to Distill Dataset Continually for Recommender Systems [69.68271683922536]
  Deep recommender models are trained under a continual learning paradigm, relying on massive and continuously growing streaming behavioral logs.<n>In large-scale platforms, retraining models on full historical data for architecture comparison or iteration is prohibitively expensive, severely slowing down model development.<n>We formulate this problem as emphstreaming dataset distillation for recommender systems and propose textbfDIET, a unified framework that maintains a compact distilled dataset which evolves alongside streaming data while preserving training-critical signals.
  arXiv  Detail & Related papers  (2026-03-26T02:50:13Z)
• DDTime: Dataset Distillation with Spectral Alignment and Information Bottleneck for Time-Series Forecasting [28.005308500582405]
  Time-series forecasting is fundamental across many domains, yet training accurate models often requires large-scale datasets and computational resources.<n>We propose DDTime, a lightweight and plug-in distillation framework built upon first-order condensation decomposition.<n> DDTime consistently outperforms existing distillation methods, achieving about 30% relative accuracy gains while introducing about 2.49% computational overhead.
  arXiv  Detail & Related papers  (2025-11-20T16:50:09Z)
• Dataset Distillation as Pushforward Optimal Quantization [2.5892916589735457]
  We propose a synthetic training set that achieves similar performance to training on real data, with orders of magnitude less computational requirements.<n>In particular, we link existing disentangled dataset distillation methods to the classical optimal quantization and Wasserstein barycenter problems.<n>We achieve better performance and inter-model generalization on the ImageNet-1K dataset with trivial additional computation, and SOTA performance in higher image-per-class settings.
  arXiv  Detail & Related papers  (2025-01-13T20:41:52Z)
• Enhancing Foundation Models for Time Series Forecasting via Wavelet-based Tokenization [74.3339999119713]
  We develop a wavelet-based tokenizer that allows models to learn complex representations directly in the space of time-localized frequencies.<n>Our method first scales and decomposes the input time series, then thresholds and quantizes the wavelet coefficients, and finally pre-trains an autoregressive model to forecast coefficients for the forecast horizon.
  arXiv  Detail & Related papers  (2024-12-06T18:22:59Z)
• Hierarchical Features Matter: A Deep Exploration of Progressive Parameterization Method for Dataset Distillation [44.03611131165989]
  We propose a novel generative parameterization method dubbed Hierarchical generative Distillation (H-PD)<n>The proposed H-PD achieves a significant performance improvement under various settings with equivalent time consumption.<n>It even surpasses current generative distillation using diffusion models under extreme compression ratios IPC=1 and IPC=10.
  arXiv  Detail & Related papers  (2024-06-09T09:15:54Z)
• Generative Dataset Distillation: Balancing Global Structure and Local Details [49.20086587208214]
  We propose a new dataset distillation method that considers balancing global structure and local details. Our method involves using a conditional generative adversarial network to generate the distilled dataset.
  arXiv  Detail & Related papers  (2024-04-26T23:46:10Z)
• Dataset Condensation for Time Series Classification via Dual Domain Matching [12.317728375957717]
  We propose a novel framework named dataset textittextbfCondensation for textittextbfTime textittextbfSeries textittextbfClassification via Dual Domain Matching. Our proposed framework aims to generate a condensed dataset that matches the surrogate objectives in both the time and frequency domains.
  arXiv  Detail & Related papers  (2024-03-12T02:05:06Z)
• Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain [54.67888148566323]
  We introduce three large-scale time series forecasting datasets from the cloud operations domain. We show it is a strong zero-shot baseline and benefits from further scaling, both in model and dataset size. Accompanying these datasets and results is a suite of comprehensive benchmark results comparing classical and deep learning baselines to our pre-trained method.
  arXiv  Detail & Related papers  (2023-10-08T08:09:51Z)
• Improved Distribution Matching for Dataset Condensation [91.55972945798531]
  We propose a novel dataset condensation method based on distribution matching. Our simple yet effective method outperforms most previous optimization-oriented methods with much fewer computational resources.
  arXiv  Detail & Related papers  (2023-07-19T04:07:33Z)

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.