面向时间序列预测的上下文内补全
In-Context Inpainting for Time Series Forecasting
查看机构详情
- Deakin University(迪肯大学)
- Applied Artificial Intelligence Initiative(应用人工智能计划)
机构由 AI 辅助整理,请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
研究人员提出ICI-Time框架,将时间序列预测转为视觉补全任务,利用大型视觉模型,在多领域实验中表现出竞争力与数据有限时的良好适应性,搭建了时间与视觉域的新范式。
中文摘要 AI 辅助
我们提出ICI-Time,这是一种新颖的框架,它将时间序列预测重新定义为视觉补全任务,利用大型视觉模型(LVM)的泛化能力。与需要专用时间架构和大量特定领域训练的方法不同,ICI-Time将时间序列转换为结构化视觉表示(面积图),并应用视觉上下文内学习,将预测重新定义为网格结构提示内的模式补全,预训练的视觉变换器无需微调或修改架构即可解决该任务。时间依赖关系通过空间布局表示,在数值域和视觉域之间存在一致且可逆的映射。在流行病学、气象学和电力系统领域进行的大量实验表明,ICI-Time的性能与深度学习基线相比具有竞争力,并且在数据有限的设置下表现出良好的适应性,引入了一种连接时间域和视觉域的新范式。
英文摘要
We propose ICI-Time, a novel framework that reframes time series forecasting as a visual inpainting task, leveraging the generalisation power of large vision models (LVMs). Unlike methods that require specialised temporal architectures and extensive domain-specific training, ICI-Time transforms time series into structured visual representations (area charts) and applies visual in-context learning, reformulating forecasting as pattern completion within a grid-structured prompt that pre-trained vision transformers can solve without fine-tuning or architectural modification. Temporal dependencies are represented through spatial layout, with a consistent, invertible mapping between numerical and visual domains. Extensive experiments across epidemiology, meteorology, and power systems demonstrate that ICI-Time performs competitively against deep learning baselines and shows promising adaptability under limited-data settings, introducing a new paradigm that bridges temporal and visual domains.