{"ID":2858685,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.06840","arxiv_id":"2510.06840","title":"CNN-TFT explained by SHAP with multi-head attention weights for time series forecasting","abstract":"Convolutional neural networks (CNNs) and transformer architectures offer strengths for modeling temporal data: CNNs excel at capturing local patterns and translational invariances, while transformers effectively model long-range dependencies via self-attention. This paper proposes a hybrid architecture integrating convolutional feature extraction with a temporal fusion transformer (TFT) backbone to enhance multivariate time series forecasting. The CNN module first applies a hierarchy of one-dimensional convolutional layers to distill salient local patterns from raw input sequences, reducing noise and dimensionality. The resulting feature maps are then fed into the TFT, which applies multi-head attention to capture both short- and long-term dependencies and to weigh relevant covariates adaptively. We evaluate the CNN-TFT on a hydroelectric natural flow time series dataset. Experimental results demonstrate that CNN-TFT outperforms well-established deep learning models, with a mean absolute percentage error of up to 2.2%. The explainability of the model is obtained by a proposed Shapley additive explanations with multi-head attention weights (SHAP-MHAW). Our novel architecture, named CNN-TFT-SHAP-MHAW, is promising for applications requiring high-fidelity, multivariate time series forecasts, being available for future analysis at https://github.com/SFStefenon/CNN-TFT-SHAP-MHAW .","short_abstract":"Convolutional neural networks (CNNs) and transformer architectures offer strengths for modeling temporal data: CNNs excel at capturing local patterns and translational invariances, while transformers effectively model long-range dependencies via self-attention. This paper proposes a hybrid architecture integrating conv...","url_abs":"https://arxiv.org/abs/2510.06840","url_pdf":"https://arxiv.org/pdf/2510.06840v1","authors":"[\"Stefano F. Stefenon\",\"João P. Matos-Carvalho\",\"Valderi R. Q. Leithardt\",\"Kin-Choong Yow\"]","published":"2025-10-08T10:08:28Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Transformer\",\"Convolutional Neural Network\"]","has_code":false,"code_links":[{"ID":608569,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2858685,"paper_url":"https://arxiv.org/abs/2510.06840","paper_title":"CNN-TFT explained by SHAP with multi-head attention weights for time series forecasting","repo_url":"https://github.com/SFStefenon/CNN-TFT-SHAP-MHAW","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}