dorsal/arxiv
View SchemaMeta-learning to Address Data Shift in Time Series Classification
| Authors | Samuel Myren, Nidhi Parikh, Natalie Klein |
|---|---|
| Categories | |
| ArXiv ID | 2601.09018vv1 |
| URL | https://arxiv.org/abs/2601.09018 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Across engineering and scientific domains, traditional deep learning (TDL) models perform well when training and test data share the same distribution. However, the dynamic nature of real-world data, broadly termed \textit{data shift}, renders TDL models prone to rapid performance degradation, requiring costly relabeling and inefficient retraining. Meta-learning, which enables models to adapt quickly to new data with few examples, offers a promising alternative for mitigating these challenges. Here, we systematically compare TDL with fine-tuning and optimization-based meta-learning algorithms to assess their ability to address data shift in time-series classification. We introduce a controlled, task-oriented seismic benchmark (SeisTask) and show that meta-learning typically achieves faster and more stable adaptation with reduced overfitting in data-scarce regimes and smaller model architectures. As data availability and model capacity increase, its advantages diminish, with TDL with fine-tuning performing comparably. Finally, we examine how task diversity influences meta-learning and find that alignment between training and test distributions, rather than diversity alone, drives performance gains. Overall, this work provides a systematic evaluation of when and why meta-learning outperforms TDL under data shift and contributes SeisTask as a benchmark for advancing adaptive learning research in time-series domains.
{
"annotation_id": "4601623a-c71e-4c27-985c-98b4deae924f",
"date_created": "2026-02-17T05:53:20.474000Z",
"date_modified": "2026-02-17T05:53:20.474000Z",
"file_hash": "e131844b23f71f39f5be63c5218e58fbd8c648eb5b18545d6361c2bb35fa0e53",
"private": false,
"record": {
"abstract": "Across engineering and scientific domains, traditional deep learning (TDL) models perform well when training and test data share the same distribution. However, the dynamic nature of real-world data, broadly termed \\textit{data shift}, renders TDL models prone to rapid performance degradation, requiring costly relabeling and inefficient retraining. Meta-learning, which enables models to adapt quickly to new data with few examples, offers a promising alternative for mitigating these challenges. Here, we systematically compare TDL with fine-tuning and optimization-based meta-learning algorithms to assess their ability to address data shift in time-series classification. We introduce a controlled, task-oriented seismic benchmark (SeisTask) and show that meta-learning typically achieves faster and more stable adaptation with reduced overfitting in data-scarce regimes and smaller model architectures. As data availability and model capacity increase, its advantages diminish, with TDL with fine-tuning performing comparably. Finally, we examine how task diversity influences meta-learning and find that alignment between training and test distributions, rather than diversity alone, drives performance gains. Overall, this work provides a systematic evaluation of when and why meta-learning outperforms TDL under data shift and contributes SeisTask as a benchmark for advancing adaptive learning research in time-series domains.",
"arxiv_id": "2601.09018",
"authors": [
"Samuel Myren",
"Nidhi Parikh",
"Natalie Klein"
],
"categories": [
"cs.LG",
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "Meta-learning to Address Data Shift in Time Series Classification",
"url": "https://arxiv.org/abs/2601.09018",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "84e3cf1f-3ade-4a27-80f0-def07f353b70",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}