AdaPTS: Adapting Univariate Foundation Models to Probabilistic Multivariate Time Series Forecasting

TL;DR

AdaPTS introduces adapter modules that enable pre-trained univariate time series models to effectively handle multivariate forecasting tasks, improving accuracy and uncertainty quantification in practical applications.

Contribution

The paper proposes a novel adapter-based framework that adapts univariate foundation models for multivariate time series forecasting, addressing feature dependencies and uncertainty estimation.

Findings

Significant improvements in forecasting accuracy on synthetic and real datasets.

Enhanced uncertainty quantification compared to baseline methods.

Demonstrated scalability and modularity of the adapter approach.

Abstract

Pre-trained foundation models (FMs) have shown exceptional performance in univariate time series forecasting tasks. However, several practical challenges persist, including managing intricate dependencies among features and quantifying uncertainty in predictions. This study aims to tackle these critical limitations by introducing adapters; feature-space transformations that facilitate the effective use of pre-trained univariate time series FMs for multivariate tasks. Adapters operate by projecting multivariate inputs into a suitable latent space and applying the FM independently to each dimension. Inspired by the literature on representation learning and partially stochastic Bayesian neural networks, we present a range of adapters and optimization/inference strategies. Experiments conducted on both synthetic and real-world datasets confirm the efficacy of adapters, demonstrating substantial enhancements in forecasting accuracy and uncertainty quantification compared to baseline methods. Our framework, AdaPTS, positions adapters as a modular, scalable, and effective solution for leveraging time series FMs in multivariate contexts, thereby promoting their wider adoption in real-world applications. We release the code at https://github.com/abenechehab/AdaPTS.