arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2409.15367cs.LGcs.AIcs.CL
收藏

使用 Wasserstein 损失微调时间序列基础模型

Fine-Tuning a Time Series Foundation Model with Wasserstein Loss

Andrei Chernov

AI总结:

针对LLM架构在时间序列预测中交叉熵损失未考虑类间距离的局限,提出使用Wasserstein损失进行微调,并在22个零样本数据集上验证了该方法显著提升了点估计性能。

摘要:

受大型语言模型(LLMs)在自然语言处理(NLP)中最新进展的启发,专注于开发时间序列预测基础模型的研究激增。一种方法是在分词后的时间序列数据上,使用交叉熵损失训练 LLM 架构。尽管该方法展现出可喜的结果,但交叉熵损失主要为分类任务设计,未考虑类间距离。为解决该局限,我们提议在此类架构中使用 Wasserstein 损失。为验证该方法,我们在 $22$ 个零样本数据集上微调了一个时间序列基础模型,比较了交叉熵损失与 Wasserstein 损失的性能。结果表明,用 Wasserstein 损失替换交叉熵损失能显著改善点估计。

英文摘要:

Inspired by recent advancements in large language models (LLMs) for Natural Language Processing (NLP), there has been a surge in research focused on developing foundational models for time series forecasting. One approach involves training LLM architectures on tokenized time series data using cross-entropy loss. Although this method has demonstrated promising results, cross-entropy loss is primarily designed for classification tasks and does not account for the distance between classes. To address this limitation, we propose using the Wasserstein loss for such architectures. To validate our approach, we fine-tuned a foundational time series model on $22$ zero-shot datasets, comparing the performance of cross-entropy loss with that of Wasserstein loss. Our results demonstrate that replacing cross-entropy loss with Wasserstein loss significantly improves point estimation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑