Neural forecasting at scale

TL;DR

This paper introduces N-BEATS(P), a scalable, parallel version of the N-BEATS model that significantly reduces training time and memory usage for large-scale multi-step time series forecasting without sacrificing accuracy.

Contribution

The paper presents N-BEATS(P), a novel global parallel model that enables efficient training of multiple univariate time series forecasting models simultaneously.

Findings

Training time reduced by 50%

Memory requirement decreased by a factor of 5

Maintains accuracy across various forecasting scenarios

Abstract

We study the problem of efficiently scaling ensemble-based deep neural networks for multi-step time series (TS) forecasting on a large set of time series. Current state-of-the-art deep ensemble models have high memory and computational requirements, hampering their use to forecast millions of TS in practical scenarios. We propose N-BEATS(P), a global parallel variant of the N-BEATS model designed to allow simultaneous training of multiple univariate TS forecasting models. Our model addresses the practical limitations of related models, reducing the training time by half and memory requirement by a factor of 5, while keeping the same level of accuracy in all TS forecasting settings. We have performed multiple experiments detailing the various ways to train our model and have obtained results that demonstrate its capacity to generalize in various forecasting conditions and setups.