{"slug": "forking-sequences-part-i-statistically-and-computationally-efficient-multi", "title": "Forking-Sequences — Part I: Statistically and Computationally Efficient Multi-Horizon Forecasting", "summary": "A new method called forking-sequences, developed by researchers including W. Potosnak and B. Oreshkin, improves multi-horizon forecasting by encoding and decoding the entire series across all forecast creation dates in a single forward pass, requiring no new parameters. Validated across six encoder types, it improves sCRPS by up to 49.3% for LSTM encoders and reduces cross-validation complexity by a factor of T, making inference order-of-magnitude cheaper.", "body_md": "Based on: Potosnak, W., Wolff, M., Cao, M., Ma, R., Konstantinova, T., Efimov, D., Mahoney, M.W., Oreshkin, B., & Olivares, K.G. \"Forking-Sequences: Statistically and Computationally Efficient Multi-Horizon Forecasting with Reduced Volatility.\" Transactions on Machine Learning Research, 2026. (Disclaimer: Code implementation not used in the paper; not affiliated with Amazon — provided as a reference for forking-sequences and forecast ensembling) TL;DR One architectural swap, no new parameters. Instead of encoding one forecast window at a time in the window-sampling training paradigm, forking-sequences encodes and decodes the entire series across all forecast creation dates in a single forward pass. Encoder-Agnostic. Validated across six encoder types — MLP, RNN, LSTM, CNN, Transformer, and State Space. Better gradients, faster training. Loss is pooled across every forecast creation date, cutting gradient variance at a linear rate and fixing vanishing gradients in RNNs/LSTMs — sCRPS improved up to 49.3% for LSTM encoders. Order-of-magnitude cheaper inference. Reusing encoder outputs across T forecast dates cuts cross-validation complexity by a factor of T for every encoder type — e.g. for LSTMs, RNNs, and CNNs, O(T²) drops to O(T). Multi-horizon forecasting has become the default approach across energy, finance, supply chain, and healthcare applications, because it gives planners visibility into […]", "url": "https://wpnews.pro/news/forking-sequences-part-i-statistically-and-computationally-efficient-multi", "canonical_source": "https://blog.ml.cmu.edu/2026/08/10/forking-sequences-part-i-statistically-and-computationally-efficient-multi-horizon-forecasting/", "published_at": "2026-08-10 21:51:11+00:00", "updated_at": "2026-08-10 21:51:20.968457+00:00", "lang": "en", "topics": ["machine-learning", "artificial-intelligence"], "entities": ["Potosnak, W.", "Wolff, M.", "Cao, M.", "Ma, R.", "Konstantinova, T.", "Efimov, D.", "Mahoney, M.W.", "Oreshkin, B."], "alternates": {"html": "https://wpnews.pro/news/forking-sequences-part-i-statistically-and-computationally-efficient-multi", "markdown": "https://wpnews.pro/news/forking-sequences-part-i-statistically-and-computationally-efficient-multi.md", "text": "https://wpnews.pro/news/forking-sequences-part-i-statistically-and-computationally-efficient-multi.txt", "jsonld": "https://wpnews.pro/news/forking-sequences-part-i-statistically-and-computationally-efficient-multi.jsonld"}}