Metraipolis › archive › dc_a5c10307d596
Halo: Improving forecast accuracy through heteroscedastic estimation — arXiv:2609.10589
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.10589
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T05:54:44.194Z
- SHA-256 of the archived text
72d42eb89ced4859438cf180be73f9c70c9e8318bd76ac0d07dbbdab36314c99- Stable URL of this page
https://metraipolis.com/documento/dc_a5c10307d596
How to cite
Halo: Improving forecast accuracy through heteroscedastic estimation — arXiv:2609.10589 — arXiv (en), https://arxiv.org/abs/2609.10589 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a5c10307d596 (Metraipolis archive dc_a5c10307d596, sha256 72d42eb89ced4859438cf180be73f9c70c9e8318bd76ac0d07dbbdab36314c99)
BibTeX
@misc{dca5c10307d596,
title = {Halo: Improving forecast accuracy through heteroscedastic estimation — arXiv:2609.10589},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.10589},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a5c10307d596; sha256 72d42eb89ced4859438cf180be73f9c70c9e8318bd76ac0d07dbbdab36314c99},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10589. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Heteroscedastic forecasting, where a network estimates a scale parameter alongside a location parameter, is normally motivated by uncertainty quantification. This paper shows it also improves the point estimate, in contrast to reported negative results for heteroscedastic estimation outside time series. Halo is a modification that reuses an existing deep forecaster's architecture, giving it a second output for the scale of its implied distribution and training it under the matching negative log likelihood. Adapting three state-of-the-art models --- a transformer, a graph network paired with a variational autoencoder, and a single-layer convolutional network --- under both Gaussian and Laplacian losses demonstrates the phenomenon. On the five electricity price markets of a standard forecasting benchmark, Halo improves MSE and MAE in 28 of 30 model-market-metric comparisons, cutting average MSE by 2.6% to 16.5% and average MAE by 1.7% to 11.0%. Two findings emerge: (1) whether the scale estimate comes from a second projection head or from a full parallel network matters far less than whether the network estimates scale, and (2) the improvement holds under the hyperparameters already tuned for the point-estimate baseline, so retuning is optional. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.10589 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-16T05:44:24.932Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).