TimeFound用多尺度分块实现跨领域零样本时间序列预测
TimeFound: A Foundation Model for Time Series Forecasting
- 采用多分辨率分块策略捕捉不同尺度的时间模式
- 在200M和710M参数规模下预训练，覆盖真实与合成数据
- 无需微调即可在多种领域和预测时长上表现优异
我们提出TimeFound，一种基于编码器-解码器变换器的时间序列基础模型，支持开箱即用的零样本预测。为处理来自不同领域的时序数据，TimeFound采用多分辨率分块策略，以捕捉复杂的时间模式。我们在一个包含真实世界和合成数据的大规模时序语料库上，对两个规模（200M 和 710M 参数）的模型进行预训练。在一系列未见数据集上的实证评估表明，TimeFound在多个领域和预测时长下，零样本预测性能优于或媲美现有最先进的时间序列基础模型。
原文摘要 · Abstract (English)
We present TimeFound, an encoder-decoder transformer-based time series foundation model for out-of-the-box zero-shot forecasting. To handle time series data from various domains, TimeFound employs a multi-resolution patching strategy to capture complex temporal patterns at multiple scales. We pre-train our model with two sizes (200M and 710M parameters) on a large time-series corpus comprising both real-world and synthetic datasets. Over a collection of unseen datasets across diverse domains and forecasting horizons, our empirical evaluations suggest that TimeFound can achieve superior or competitive zero-shot forecasting performance, compared to state-of-the-art time series foundation models.
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。