N-BEATS: Neural basis expansion analysis for interpretable time series forecasting
Reviewed by Pithpith:472JKLW2open to challenge →
read the original abstract
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.
This paper has not been read by Pith yet.
Forward citations
Cited by 25 Pith papers
-
Stabilizing distribution-free probabilistic forecasts
Neural network-parameterized regression splines enable joint optimization of forecast quality and stability in distribution-free probabilistic time series models by penalizing dissimilarities from forecast updates.
-
C2L-Net: A Data-Driven Model for State-of-Charge Estimation of Lithium-Ion Batteries During Discharge
C2L-Net delivers competitive SOC estimation accuracy on drive-cycle data with up to 60x faster inference by using chunk-based attention, Fourier seasonality, causal GRU encoding, and a recursive-style latest decoder.
-
Sundial: A Family of Highly Capable Time Series Foundation Models
Sundial uses TimeFlow Loss for native pre-training of Transformers on continuous time series from TimeBench, achieving SOTA point and probabilistic forecasting with millisecond inference.
-
Deep Time Series Models: A Comprehensive Survey and Benchmark
This survey and benchmark of deep time series models using the released TSLib library finds that models with specific structures perform well only on distinct analysis tasks.
-
Timesynth: A Temporal Fidelity Framework for Health Signal Digital Twins
TimeSynth supplies a generator and fidelity diagnostics that reveal pointwise metrics miss up to 53° phase errors across 11 architectures and show architecture determines temporal fidelity in health signals.
-
Deployment-Side Adaptiveness in Multi-Horizon Volatility Forecasting
Validation-based selection of inference-time rollout rules for multi-output volatility forecasters yields low-cost improvements over default MIMO deployment and recovers much of ensemble benefit at lower cost.
-
Mechanical Field Networks: Structured Neural Dynamics for Multivariate Systems
MF-Net learns a shared field state and mechanical transition rule from trajectories to deliver competitive forecasting and recoverable relation matrices on Lorenz-96 and real systems.
-
Fast Training of Mixture-of-Experts for Time Series Forecasting via Expert Loss Integration
A new MoE training method integrates expert-level losses and partial online updates to improve forecasting accuracy and efficiency over standard statistical and neural models.
-
A Foundation Model for Instruction-Conditioned In-Context Time Series Tasks
iAmTime is a time-series foundation model that uses instruction-conditioned in-context learning from demonstrations to perform zero-shot adaptation on forecasting, imputation, classification, and related tasks.
-
ReNF: Rethinking the Design of Neural Long-Term Time Series Forecasters
ReNF proposes Boosted Direct Output (BDO) and parameter smoothing so a basic temporal MLP outperforms complex state-of-the-art models on long-term time series forecasting benchmarks by implicitly combining forecasts t...
-
An AI system to help scientists write expert-level empirical software
ERA combines LLMs and tree search to produce expert-level empirical software that outperforms top human methods on single-cell analysis leaderboards and CDC COVID-19 forecasts.
-
An AI system to help scientists write expert-level empirical software
ERA is an AI system using LLMs and tree search to produce expert-level empirical software, generating methods that outperformed top human approaches in single-cell data analysis and COVID-19 forecasting tasks.
-
Zeus: Towards Tuning-Free Foundation Model for Time Series Analysis
Zeus proposes a multi-scale Transformer with point-wise tokenization and Multi-Objective Temporal Masking to enable tuning-free performance on forecasting, interpolation, and other time series tasks.
-
Pretrained Time-Series Foundation Models for Financial Return Forecasting
Pretrained TSFMs achieve top ranks on equity return tasks but show sparse, minimal improvements over random walk, serving as practical priors without reliable alpha generation.
-
Speaking Numbers to LLMs: Multi-Wavelet Number Embeddings for Time Series Forecasting
TempoWave maps scalar observations to multi-wavelet multi-scale digit embeddings that override standard LLM tokens and improve forecasting performance on five context-enriched benchmarks to a new state-of-the-art.
-
Benchmarking Deep Time Series Models for Equity Portfolios
Benchmark of 15 time-series architectures on equity portfolios finds no model dominates, with TransEnc-8 at 0.352 rank-1 acceptability and all promoted models showing negative net Sharpe at 20 bps costs under constraints.
-
Kalimati Vegetable Price Index Forecasting with a Momentum Corrected Online Stacking Ensemble
A momentum-corrected online stacking ensemble forecasts the new Kalimati Vegetable Price Index with RMSE 1.771, MAPE 0.68 percent, and R-squared 0.845 at the 90-day horizon.
-
TSNN: A Non-parametric and Interpretable Framework for Traffic Time Series Forecasting
TSNN matches time series entries to a training-derived memory bank to forecast traffic without any trainable parameters and achieves competitive accuracy on four real-world datasets.
-
CollideNet: Hierarchical Multi-scale Video Representation Learning with Disentanglement for Time-To-Collision Forecasting
CollideNet achieves state-of-the-art time-to-collision forecasting on three public datasets by combining multi-scale spatial aggregation with temporal disentanglement of trend and seasonality in a hierarchical transformer.
-
Degradation-aware Predictive Energy Management for Fuel Cell-Battery Ship Power System with Data-driven Load Forecasting
A degradation-aware predictive controller for hybrid ship power systems reduces hydrogen consumption by up to 5.8% and fuel cell degradation by up to 36.4% versus a filter-based benchmark on real harbor tug data.
-
A Foundation Model for Instruction-Conditioned In-Context Time Series Tasks
iAmTime is a hierarchical transformer-based time series foundation model that uses semantic tokens and instruction-conditioned prompts to infer tasks from demonstrations, achieving improved zero-shot performance on fo...
-
One Step Closer to Ground Truth: A Multi-Scale Residual-Aware Representation Learning Pipeline for Predicting Time Series Data
A two-stage residual-aware framework adds a meta-corrector after a base transformer to model structured errors and reports state-of-the-art results on eight time-series benchmarks.
-
Graph Convolutional Support Vector Regression for Robust Spatiotemporal Forecasting of Urban Air Pollution
GCSVR combines graph convolutions for spatial station dependencies with SVR for nonlinear temporal patterns, yielding more accurate and stable air pollution forecasts on Delhi and Mumbai datasets than standard benchmarks.
-
Machine Learning-Based Characterization of Solar p-Mode Frequency Shifts during Solar Cycle 25
Machine learning and time-series methods are applied to characterize solar p-mode frequency shifts for solar cycle 25 as a potential early indicator of solar activity.
-
Time Series Analysis in Machine Learning
A review chapter covering basic time series concepts, classical models like ARIMA, and ML approaches including neural networks and transformers.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.