UniCL: A Universal Contrastive Learning Framework for Large Time Series Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Jiawei, Peng, Jingshu, Li, Haoyang, Chen, Lei
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914799968845824
author Li, Jiawei
Peng, Jingshu
Li, Haoyang
Chen, Lei
author_facet Li, Jiawei
Peng, Jingshu
Li, Haoyang
Chen, Lei
contents Time-series analysis plays a pivotal role across a range of critical applications, from finance to healthcare, which involves various tasks, such as forecasting and classification. To handle the inherent complexities of time-series data, such as high dimensionality and noise, traditional supervised learning methods first annotate extensive labels for time-series data in each task, which is very costly and impractical in real-world applications. In contrast, pre-trained foundation models offer a promising alternative by leveraging unlabeled data to capture general time series patterns, which can then be fine-tuned for specific tasks. However, existing approaches to pre-training such models typically suffer from high-bias and low-generality issues due to the use of predefined and rigid augmentation operations and domain-specific data training. To overcome these limitations, this paper introduces UniCL, a universal and scalable contrastive learning framework designed for pretraining time-series foundation models across cross-domain datasets. Specifically, we propose a unified and trainable time-series augmentation operation to generate pattern-preserved, diverse, and low-bias time-series data by leveraging spectral information. Besides, we introduce a scalable augmentation algorithm capable of handling datasets with varying lengths, facilitating cross-domain pretraining. Extensive experiments on two benchmark datasets across eleven domains validate the effectiveness of UniCL, demonstrating its high generalization on time-series analysis across various fields.
format Preprint
id arxiv_https___arxiv_org_abs_2405_10597
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle UniCL: A Universal Contrastive Learning Framework for Large Time Series Models
Li, Jiawei
Peng, Jingshu
Li, Haoyang
Chen, Lei
Machine Learning
Artificial Intelligence
Computation and Language
Time-series analysis plays a pivotal role across a range of critical applications, from finance to healthcare, which involves various tasks, such as forecasting and classification. To handle the inherent complexities of time-series data, such as high dimensionality and noise, traditional supervised learning methods first annotate extensive labels for time-series data in each task, which is very costly and impractical in real-world applications. In contrast, pre-trained foundation models offer a promising alternative by leveraging unlabeled data to capture general time series patterns, which can then be fine-tuned for specific tasks. However, existing approaches to pre-training such models typically suffer from high-bias and low-generality issues due to the use of predefined and rigid augmentation operations and domain-specific data training. To overcome these limitations, this paper introduces UniCL, a universal and scalable contrastive learning framework designed for pretraining time-series foundation models across cross-domain datasets. Specifically, we propose a unified and trainable time-series augmentation operation to generate pattern-preserved, diverse, and low-bias time-series data by leveraging spectral information. Besides, we introduce a scalable augmentation algorithm capable of handling datasets with varying lengths, facilitating cross-domain pretraining. Extensive experiments on two benchmark datasets across eleven domains validate the effectiveness of UniCL, demonstrating its high generalization on time-series analysis across various fields.
title UniCL: A Universal Contrastive Learning Framework for Large Time Series Models
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2405.10597