In-Context Learning Implicit Representation for Time Series Forecasting

Liwei Yang

doi:10.1109/LSP.2025.3624128

In-Context Learning Implicit Representation for Time Series Forecasting

2025-01-01

0

PRE

AI

DOI：10.1109/LSP.2025.3624128

Original

Original request for help

Save

Abstract

En 中文

Time series forecasting task aims at predicting future time series signal given historical time series observations. This letter investigates time series signal modeling and forecasting from a novel perspective named In-Context Learning Implicit Representation (ICL-IR). ICL-IR formulates the time series data as an Implicit Representation (IR), which is a mapping from timestamps to time series data. By incorporating the decoder-only transformer architecture, ICL-IR automatically identifies the mapping from timestamps to data based on the constructed in-context examples and predicts future time series data in an autoregressive fashion. To improve the autoregressive prediction performance, we design a One-Step-ahead Training (OST) strategy and demonstrate its effectiveness through both theoretical analysis and empirical evaluation. Experiments demonstrate the superiority of ICL-IR for the time series forecasting task.

Keywords:

Dynamics modeling

implicit representation

in-context learning

time series forecasting

In-Context Learning Implicit Representation for Time Series Forecasting

Abstract

Journal

Organization

Cited Papers