Bridging Distribution Gaps in Time Series Foundation Model Pretraining with Prototype-Guided Normalization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gong, Peiliang, Eldele, Emadeldeen, Wu, Min, Chen, Zhenghua, Li, Xiaoli, Zhang, Daoqiang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909580040077312
author Gong, Peiliang
Eldele, Emadeldeen
Wu, Min
Chen, Zhenghua
Li, Xiaoli
Zhang, Daoqiang
author_facet Gong, Peiliang
Eldele, Emadeldeen
Wu, Min
Chen, Zhenghua
Li, Xiaoli
Zhang, Daoqiang
contents Foundation models have achieved remarkable success across diverse machine-learning domains through large-scale pretraining on large, diverse datasets. However, pretraining on such datasets introduces significant challenges due to substantial mismatches in data distributions, a problem particularly pronounced with time series data. In this paper, we tackle this issue by proposing a domain-aware adaptive normalization strategy within the Transformer architecture. Specifically, we replace the traditional LayerNorm with a prototype-guided dynamic normalization mechanism (ProtoNorm), where learned prototypes encapsulate distinct data distributions, and sample-to-prototype affinity determines the appropriate normalization layer. This mechanism effectively captures the heterogeneity of time series characteristics, aligning pretrained representations with downstream tasks. Through comprehensive empirical evaluation, we demonstrate that our method significantly outperforms conventional pretraining techniques across both classification and forecasting tasks, while effectively mitigating the adverse effects of distribution shifts during pretraining. Incorporating ProtoNorm is as simple as replacing a single line of code. Extensive experiments on diverse real-world time series benchmarks validate the robustness and generalizability of our approach, advancing the development of more versatile time series foundation models.
format Preprint
id arxiv_https___arxiv_org_abs_2504_10900
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bridging Distribution Gaps in Time Series Foundation Model Pretraining with Prototype-Guided Normalization
Gong, Peiliang
Eldele, Emadeldeen
Wu, Min
Chen, Zhenghua
Li, Xiaoli
Zhang, Daoqiang
Machine Learning
Artificial Intelligence
Foundation models have achieved remarkable success across diverse machine-learning domains through large-scale pretraining on large, diverse datasets. However, pretraining on such datasets introduces significant challenges due to substantial mismatches in data distributions, a problem particularly pronounced with time series data. In this paper, we tackle this issue by proposing a domain-aware adaptive normalization strategy within the Transformer architecture. Specifically, we replace the traditional LayerNorm with a prototype-guided dynamic normalization mechanism (ProtoNorm), where learned prototypes encapsulate distinct data distributions, and sample-to-prototype affinity determines the appropriate normalization layer. This mechanism effectively captures the heterogeneity of time series characteristics, aligning pretrained representations with downstream tasks. Through comprehensive empirical evaluation, we demonstrate that our method significantly outperforms conventional pretraining techniques across both classification and forecasting tasks, while effectively mitigating the adverse effects of distribution shifts during pretraining. Incorporating ProtoNorm is as simple as replacing a single line of code. Extensive experiments on diverse real-world time series benchmarks validate the robustness and generalizability of our approach, advancing the development of more versatile time series foundation models.
title Bridging Distribution Gaps in Time Series Foundation Model Pretraining with Prototype-Guided Normalization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2504.10900