Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

interpretable time series forecasting

Interpretable time series forecasting is the process of predicting future values in sequential, time-ordered data using models designed to provide clear, human-understandable explanations of their predictions. Unlike black-box machine learning approaches, interpretable forecasting systems reveal how specific inputs and temporal dynamics drive the forecasted results. This transparency is typically achieved by decomposing predictions into recognizable time series components, such as trend and seasonality, or by explicitly attributing contributions to historical observations, lagged variables, and external factors. By illuminating the internal logic behind the numbers, this approach enables practitioners to validate forecast reliability, diagnose model errors, and make domain-informed decisions with greater confidence.

1 item

N-BEATS: Neural basis expansion analysis for interpretable time series forecasting

N-BEATS: Neural basis expansion analysis for interpretable time series forecasting

Boris N. Oreshkin, Dmitri Carpov, Nicolas Chapados, Yoshua Bengio

OrganizationsElement AIMila

Why you should read this

Proposes a pure deep learning architecture comprised of backward and forward residual basis expansions that matches classical interpretability without requiring domain-specific feature engineering.

We focus on solving the univariate times series point forecasting problem using deep learning. We propose a deep neural architecture based on backward and forward residual links and a very deep stack of fully-connected layers. The architecture has a number of desirable properties, being interpretable, applicable without modification to a wide array of target domains, and fast to train. We test the proposed architecture on several well-known datasets, including M3, M4 and TOURISM competition datasets containing time series from diverse domains. We demonstrate state-of-the-art performance for two configurations of N-BEATS for all the datasets, improving forecast accuracy by 11% over a statistical benchmark and by 3% over last year's winner of the M4 competition, a domain-adjusted hand-crafted hybrid between neural network and statistical time series models. The first configuration of our model does not employ any time-series-specific components and its performance on heterogeneous datasets strongly suggests that, contrarily to received wisdom, deep learning primitives such as residual blocks are by themselves sufficient to solve a wide range of forecasting problems. Finally, we demonstrate how the proposed architecture can be augmented to provide outputs that are interpretable without considerable loss in accuracy.

Added

2026-06-27