通过对比课程学习提升时间序列模型在建筑能耗预测中的表现。
Enabling Time-series Foundation Model for Building Energy Forecasting via Contrastive Curriculum Learning
- 设计对比课程学习策略优化训练数据顺序。
- 零样本/少样本性能相比现有模型提升14.6%。
- 适合关注时序建模与能源预测的从业者。
时间序列预测的进步正推动传统机器学习模型向具备通用知识的时空基础模型（TSFM）转变。然而，现有基础模型在建筑能耗预测（BEF）等能源领域仍表现不佳。本文研究了基础模型在BEF任务中的适应性问题，从模型和数据两个角度揭示了直接微调的不足。为此，提出一种基于对比课程学习的新训练方法，优化了在时序基础模型适配背景下的训练数据排序。实验表明，该方法相比现有基础模型，在零样本和少样本场景下性能提升14.6%。代码与新构建的时序基础模型将公开于匿名GitHub仓库。
原文摘要 · Abstract (English)
Advances in time-series forecasting are driving a shift from conventional machine learning models to foundation models (FMs) that are trained with generalized knowledge. However, existing FMs still perform poorly in the energy fields, such as building energy forecasting (BEF). This paper studies the adaptation of FM to BEF tasks. We demonstrate the shortcomings of fine-tuning FM straightforwardly from both the perspectives of FM and the data. To overcome these limitations, we propose a new \textit{contrastive curriculum learning}-based training method. Our method optimizes the ordering of training data in the context of TSFM adaptation. Experiments show that our method can improve the zero/few-shot performance by 14.6\% compared to the existing FMs. Our code and new TSFM will be available at <Anonymous Github Repo>.
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。