Deep Factors with Gaussian Processes for Forecasting

TL;DR

This paper introduces a hybrid forecasting model combining deep neural networks and Gaussian Processes to improve accuracy and uncertainty estimation in large-scale time series data.

Contribution

The paper presents a scalable, data-driven hybrid model that integrates deep factors with Gaussian Processes for enhanced time series forecasting.

Findings

Outperforms state-of-the-art methods in accuracy

Provides reliable uncertainty estimates

Scales effectively to large datasets

Abstract

A large collection of time series poses significant challenges for classical and neural forecasting approaches. Classical time series models fail to fit data well and to scale to large problems, but succeed at providing uncertainty estimates. The converse is true for deep neural networks. In this paper, we propose a hybrid model that incorporates the benefits of both approaches. Our new method is data-driven and scalable via a latent, global, deep component. It also handles uncertainty through a local classical Gaussian Process model. Our experiments demonstrate that our method obtains higher accuracy than state-of-the-art methods.