Metraipolis › archive › dc_35150a3c8f99
Beyond Homoscedasticity: Decoupled Uncertainty Optimization for Deep Imbalanced Regression — arXiv:2609.04995
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.04995
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T22:58:27.656Z
- SHA-256 of the archived text
7ed1a65a955584724d5b3bd14799f1fb1672096ff2d358248619b3bb40fdc9b0- Stable URL of this page
https://metraipolis.com/documento/dc_35150a3c8f99
How to cite
Beyond Homoscedasticity: Decoupled Uncertainty Optimization for Deep Imbalanced Regression — arXiv:2609.04995 — arXiv (en), https://arxiv.org/abs/2609.04995 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_35150a3c8f99 (Metraipolis archive dc_35150a3c8f99, sha256 7ed1a65a955584724d5b3bd14799f1fb1672096ff2d358248619b3bb40fdc9b0)
BibTeX
@misc{dc35150a3c8f99,
title = {Beyond Homoscedasticity: Decoupled Uncertainty Optimization for Deep Imbalanced Regression — arXiv:2609.04995},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.04995},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_35150a3c8f99; sha256 7ed1a65a955584724d5b3bd14799f1fb1672096ff2d358248619b3bb40fdc9b0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04995. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Deep Imbalanced Regression (DIR) is pervasive in continuous prediction tasks across diverse modalities, such as age estimation, depth prediction, and protein mutation activity prediction, where label-scarce tail samples often carry higher practical value. However, most existing methods still learn deterministic point mappings under mean squared error or its simple variants, implicitly assuming a uniform uncertainty level across all samples and thereby overlooking the instance-wise heteroscedasticity that is widespread in long-tailed data. We further point out that even heteroscedastic negative log-likelihood suffers from a gradient coupling issue, which, under DIR scenarios, weakens the learning signal of hard tail samples and leads to optimization inertia as well as tail underfitting. To address this, we propose DUO, an uncertainty-aware long-tailed regression framework. Specifically, the proposed method models the regression target as a conditional Gaussian distribution to explicitly characterize instance-level predictive uncertainty, and transforms uncertainty into a dynamic enhancement signal for tail samples through decoupled mean-variance optimization. Furthermore, we design a distribution-guided contrastive learning mechanism that adaptively constructs positive and negative pairs based on the overlap between sample distributions, thereby alleviating feature looseness and cross-label semantic entanglement. Across visual and biological DIR benchmarks, DUO achieves the best few-shot bMAE and GM on IMDB-WIKI-DIR, AgeDB-DIR, and AAV2-DIR while remaining competitive on few-shot MAE. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.04995 — 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 2026-10-16T22:31:32.201Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).