针对时间序列模型的高效微调方法，提升长短期预测性能。
TRACE: Time SeRies PArameter EffiCient FinE-tuning
- 基于门控动态重要性计算的参数高效微调机制
- 重构预测头显著减少参数量且保持优异性能
- 适用于长周期预测与多任务场景，尤其适合资源受限环境
我们提出一种面向时间序列基础模型的高效微调方法——TRACE（Time Series Parameter Efficient Fine-tuning）。尽管预训练时间序列模型日益流行，但仍面临两大挑战：(1) 与自然语言任务不同，时间序列数据在频率、通道数、历史/预测长度上差异显著；长时序预测任务中，定制化微调可显著提升性能。(2) 现有参数高效微调方法如LoRA仍适用，但需适配时间特性。为此，我们的TRACE框架引入两项创新：(1) 门控动态模拟重要性计算（Gated DSIC），一种无偏的LoRA模块重要性选择机制，确保掩码前后条件参数一致性；实验表明其优于常规微调。(2) 针对长时序预测任务的重构预测头，在参数量大幅降低的同时，性能可媲美甚至超越线性探测头。在多种数据集上的长/短期预测、异常检测及自然语言任务的广泛实验，结合消融研究，验证了该方法的有效性。
原文摘要 · Abstract (English)
We propose an efficient fine-tuning method for time series foundation models, termed TRACE: Time Series Parameter Efficient Fine-tuning. While pretrained time series foundation models are gaining popularity, they face the following challenges: (1) Unlike natural language tasks, time series data vary in frequency, channel numbers, historical/prediction lengths. For long-term forecasting tasks in particular, tailored fine-tuning can significantly enhance performance.(2) Existing parameter-efficient tuning methods like LoRA remain applicable but require adaptation to temporal characteristics. To address these challenges, our TRACE framework introduces two key innovations: (1) Gated DSIC (Gated Dynamic Simulation Importance Calculation), an unbiased LoRA module importance selection mechanism that ensures conditional parameter consistency before and after masking. Experiments demonstrate that Gated DSIC outperforms common fine-tuning. (2) Reconstructed prediction heads for long-term forecasting tasks, which achieve comparable or superior performance to linear probing heads while drastically reducing parameter counts. Extensive experiments on long-/short-term forecasting, anomaly detection and natural language tasks across diverse datasets, coupled with ablation studies, validate the effectiveness of our method.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。