What Matters in Deep Learning for Time Series Forecasting?

• URL: http://arxiv.org/abs/2512.22702v1
• Date: Sat, 27 Dec 2025 20:50:28 GMT
• Title: What Matters in Deep Learning for Time Series Forecasting?
• Authors: Valentina Moretti, Andrea Cini, Ivan Marisca, Cesare Alippi,
• Abstract summary: We discuss the design dimensions and trade-offs that can explain, often unexpected, observed results.<n>This paper discusses the necessity of grounding model design on principles for forecasting groups of time series.<n>We show that accounting for these aspects can be more relevant for achieving accurate results than adopting specific sequence modeling layers.
• Score: 23.46740873701343
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep learning models have grown increasingly popular in time series applications. However, the large quantity of newly proposed architectures, together with often contradictory empirical results, makes it difficult to assess which components contribute significantly to final performance. We aim to make sense of the current design space of deep learning architectures for time series forecasting by discussing the design dimensions and trade-offs that can explain, often unexpected, observed results. This paper discusses the necessity of grounding model design on principles for forecasting groups of time series and how such principles can be applied to current models. In particular, we assess how concepts such as locality and globality apply to recent forecasting architectures. We show that accounting for these aspects can be more relevant for achieving accurate results than adopting specific sequence modeling layers and that simple, well-designed forecasting architectures can often match the state of the art. We discuss how overlooked implementation details in existing architectures (1) fundamentally change the class of the resulting forecasting method and (2) drastically affect the observed empirical results. Our results call for rethinking current faulty benchmarking practices and the need to focus on the foundational aspects of the forecasting problem when designing architectures. As a step in this direction, we propose an auxiliary forecasting model card, whose fields serve to characterize existing and new forecasting architectures based on key design choices.

Related papers

• Demystifying Data-Driven Probabilistic Medium-Range Weather Forecasting [63.8116386935854]
  We demonstrate that state-of-the-art probabilistic skill requires neither intricate architectural constraints nor specialized trainings.<n>We introduce a scalable framework for learning multi-scale atmospheric dynamics by combining a directly downsampled latent space with a history-conditioned local projector.<n>We find that our framework design is robust to the choice of probabilistic estimators, seamlessly supporting interpolants, diffusion models, and CRPS-based ensemble training.
  arXiv  Detail & Related papers  (2026-01-26T03:52:16Z)
• Modèles de Fondation et Ajustement : Vers une Nouvelle Génération de Modèles pour la Prévision des Séries Temporelles [26.28141834580785]
  Foundations models have been developed for zero-shot time series forecasting.<n>These models learn generalizable representations for both point and probabilistic forecasting.<n>We study the effect of fine-tuning after pretraining to enhance their performance on specific datasets.
  arXiv  Detail & Related papers  (2025-11-27T18:19:20Z)
• Understanding the Implicit Biases of Design Choices for Time Series Foundation Models [90.894232610821]
  Time series foundation models (TSFMs) are a class of potentially powerful, general-purpose tools for time series forecasting and related temporal tasks.<n>Their behavior is strongly shaped by subtle inductive biases in their design.<n>We show how these biases can be intuitive or very counterintuitive, depending on properties of the model and data.
  arXiv  Detail & Related papers  (2025-10-22T04:42:35Z)
• A Guide to Robust Generalization: The Impact of Architecture, Pre-training, and Optimization Strategy [50.33847176405776]
  We present an empirical study spanning 6 datasets, 40 pretrained architectures, 2 specialized losses, and 3 adaptation protocols.<n>We find that convolutional neural networks pretrained in a supervised manner on large datasets often perform best.
  arXiv  Detail & Related papers  (2025-08-12T19:38:35Z)
• TimeRecipe: A Time-Series Forecasting Recipe via Benchmarking Module Level Effectiveness [21.753566949357026]
  TimeRecipe is a framework that systematically evaluates time-series forecasting methods at the module level.<n>TimeRecipe conducts over 10,000 experiments to assess the effectiveness of individual components.<n>Our results reveal that exhaustive exploration of the design space can yield models that outperform existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-06-06T19:11:48Z)
• A Survey of Model Architectures in Information Retrieval [59.61734783818073]
  The period from 2019 to the present has represented one of the biggest paradigm shifts in information retrieval (IR) and natural language processing (NLP)<n>We trace the development from traditional term-based methods to modern neural approaches, particularly highlighting the impact of transformer-based models and subsequent large language models (LLMs)<n>We conclude with a forward-looking discussion of emerging challenges and future directions.
  arXiv  Detail & Related papers  (2025-02-20T18:42:58Z)
• Exploring the design space of deep-learning-based weather forecasting systems [56.129148006412855]
  This paper systematically analyzes the impact of different design choices on deep-learning-based weather forecasting systems. We study fixed-grid architectures such as UNet, fully convolutional architectures, and transformer-based models. We propose a hybrid system that combines the strong performance of fixed-grid models with the flexibility of grid-invariant architectures.
  arXiv  Detail & Related papers  (2024-10-09T22:25:50Z)
• Optimizing Time Series Forecasting Architectures: A Hierarchical Neural Architecture Search Approach [9.706293832806082]
  We propose a novel hierarchical neural architecture search approach for time series forecasting tasks.<n>With the design of a hierarchical search space, we incorporate many architecture types designed for forecasting tasks.<n>Results on long-term-time-series-forecasting tasks show that our approach can search for lightweight high-performing forecasting architectures.
  arXiv  Detail & Related papers  (2024-06-07T17:02:37Z)
• Spatiotemporal Observer Design for Predictive Learning of High-Dimensional Data [6.214987339902511]
  An observer theory-guided deep learning architecture, called Stemporal, is designed for predictive learning Observer high dimensional data. This framework could capture thetemporaltemporal dynamics make accurate predictions in both one-step and multi-step-ahead scenarios.
  arXiv  Detail & Related papers  (2024-02-23T12:28:31Z)
• Graph Deep Learning for Time Series Forecasting [25.911123797362315]
  Graph-based predictors leverage pairwise relationships by conditioning forecasts on graphs spanning the time series collection.<n>This tutorial paper aims to introduce a comprehensive methodological framework formalizing the forecasting problem and providing design principles for graph-based predictors.
  arXiv  Detail & Related papers  (2023-10-24T16:26:38Z)
• Leveraging the structure of dynamical systems for data-driven modeling [111.45324708884813]
  We consider the impact of the training set and its structure on the quality of the long-term prediction. We show how an informed design of the training set, based on invariants of the system and the structure of the underlying attractor, significantly improves the resulting models.
  arXiv  Detail & Related papers  (2021-12-15T20:09:20Z)
• Supporting Optimal Phase Space Reconstructions Using Neural Network Architecture for Time Series Modeling [68.8204255655161]
  We propose an artificial neural network with a mechanism to implicitly learn the phase spaces properties. Our approach is either as competitive as or better than most state-of-the-art strategies.
  arXiv  Detail & Related papers  (2020-06-19T21:04:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.