发表机构
Ghent University（根特大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对冷启动时间序列预测，提出时间模型无关元学习（TMAML），通过构建时间一致的元任务并优化模型，使预测模型在少数据下快速适应，实证表明其优于或匹配标准训练模型。
AI 中文摘要
冷启动预测，即在几乎没有历史数据的情况下对时间序列进行预测的任务，是一个常见的挑战。解决这一问题需要能够从少量数据点中快速学习的方法，并利用来自相关序列的信息（通常通过静态协变量）来很好地泛化到未见过的序列。虽然一些全局预测模型可以通过利用多个序列间共享的信息生成冷启动预测，但它们并未针对训练集外泛化或从短历史中进行适应进行优化。在本工作中，我们将冷启动预测表述为一个少窗口学习问题，并引入时间模型无关元学习（TMAML），该方法将最初为神经网络少样本适应而开发的模型无关元学习算法定制用于深度时间序列预测。TMAML将元任务构建为时间上一致的支持-查询窗口，并将其与时间元训练和元测试流程配对，从而产生明确针对冷启动预测优化的预测模型。我们在时间融合变换器（TFT）上实例化TMAML，并给出了在三种冷启动场景下预测准确性和校准的初步实证分析：TMAML始终优于或匹配标准ERM训练的TFT，在三种场景中的两种上产生比朴素方法更好校准的预测，但在概率预测准确性上并未始终优于朴素方法。
英文摘要
Cold-start forecasting, the task of forecasting a time series with little to no historical data, is a common challenge. Addressing it requires approaches that learn quickly from few datapoints and leverage information from related series, typically through static covariates, to generalize well to unseen series. While some global forecasting models can generate cold-start predictions by leveraging information shared across multiple series, they are not optimized for out-of-train-set generalization or adaptation from short histories. In this work, we formulate cold-start forecasting as a few-window learning problem and introduce Temporal Model-Agnostic Meta-Learning (TMAML), which tailors the model-agnostic meta-learning algorithm, originally developed for few-shot adaptation of neural networks, to deep time series forecasting. TMAML constructs meta-tasks as temporally consistent support-query windows and pairs them with a temporal meta-training and meta-testing procedure, yielding forecasting models that are explicitly optimized for cold-start forecasting. We instantiate TMAML on the Temporal Fusion Transformer (TFT) and present an initial empirical analysis of forecast accuracy and calibration across three cold-start scenarios: TMAML consistently outperforms or matches a standard ERM-trained TFT, yields better-calibrated forecasts than naive on two of the three scenarios, but does not consistently outperform naive on probabilistic forecast accuracy.
Comments13 pages, 3 figures, Accepted at Neurips 2026 TS-LIMITS Workshop
