Smart Forecasting: Deep Learning and Explainable AI for Real-World Time Series Prediction

In recent years, the application of deep learning methods for time series forecasting has grown significantly, driven by their ability to capture complex temporal dependencies. However, their adaptability to diverse, real-world scenarios is still being validated. Financial, climate, healthcare, and industrial data present unique challenges, demanding rigorous exploration of deep learning models’ performance and flexibility.

This Research Topic seeks to investigate the efficacy of deep learning, including the latest advancements in generative AI, such as transformers, large language models (LLMs), generative adversarial networks (GANs), diffusion models, and variational autoencoders (VAEs), in time series prediction across varied domains, which integrate eXplainable AI (XAI) techniques to enhance model interpretability. We aim to address the following questions:
• How do emerging architectures perform on different types of time series data? What domain-specific challenges arise when deploying these models in real-world applications?
• How can generative AI techniques be adapted or combined with traditional models to enhance predictive accuracy in time series forecasting?
• What strategies can improve the interpretability and explainability of deep learning models in complex prediction tasks? In what ways can XAI tools increase the transparency and trustworthiness of deep learning models in high-stakes domains?
• How can we manage common data issues, such as sparsity, noise, irregular sampling, and concept drift, with advanced deep learning techniques?

We invite contributions that examine deep learning models applied to time series data, addressing challenges like data quality, sampling irregularities, and evolving patterns.

This call for papers seeks contributions that relate theoretical concepts and practical implementations, including original research articles, reviews and methodological papers.

Keywords: Deep Learning, XAI, Time series analysis, Big Data Technology, Machine Learning Models

Important note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.