時系列の生成形AI

抄録

Generative Artifical Intelligence (AI) has recently become a powerful tool and has a wide range of applications. It can be used to generate various types of data such as images and text. However, the usage of generative AI with time series is an underdeveloped field. We propose using a diffusion model combined with a language model to generate time series. Through the proposed method, we demonstrate that time series can be created from text. This research has applications in data augmentation, time series forcasting, and transfer learning.