arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

Transactions on Machine Learning Research · 期刊 · Machine Learning

2026-08-26 至 2026-08-26 共收录 3
2608.24518 2026-08-26 cs.LG cs.CV 新提交

It depends: Incorporating correlations for joint aleatoric and epistemic uncertainties of high-dimensional output spaces

视情况而定:纳入高维输出空间的偶然不确定性与认知不确定性的相关性

Leonhard F. Feiner, Manuel Nickel, Martin Menten, Laurin Lux, Rickmer Braren, Daniel Rueckert, Georgios Kaissis, Raphael Rehms, Johannes Paetzold

AI总结 该研究针对高维输出空间的深度学习,提出低秩加对角协方差结构的联合不确定性建模方法,实现偶然与认知不确定性的统一,在多项视觉任务中提升了不确定性量化效果。

Comments Published in Transactions on Machine Learning Research (TMLR), 2026. 42 pages, 14 figures, 10 tables. this https URL (https://openreview.net/forum?id=zw5EuUnBny)

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.11578 2026-08-26 cs.AI 版本更新

Efficient LLM Collaboration via Planning

通过规划实现高效的LLM协作

Byeongchan Lee, Jonghoon Lee, Dongyoung Kim, Jaehyung Kim, Kyungjoon Park, Dongjun Lee, Jinwoo Shin

机构 * KAIST(韩国科学技术院) Yonsei University(延世大学)

AI总结 本文提出COPE框架,通过规划模型与执行模型的协作,提升小模型和大模型在复杂任务中的性能,同时显著降低推理成本。

Comments Accepted at TMLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.05335 2026-08-26 stat.ML cs.LG math.ST 版本更新

Bayes with No Shame: Admissibility Geometries of Predictive Inference

贝叶斯无羞耻:预测推断的可接受几何学

Nicholas G. Polson, Daniel Zantedeschi

机构 * Booth School of Business, University of Chicago(芝加哥大学商学院) Muma College of Business, University of South Florida(佛罗里达州立大学穆马商学院)

AI总结 本文探讨了四种不同的可接受几何学,用于序列和分布无关推断,提出各类可接受程序相互不嵌套,且具有不同的最优性证书。

Comments Accepted for publication in Transactions on Machine Learning Research (TMLR). 37 pages, 6 figures, 5 tables. This replacement adds author information, corrects localized scope statements and supporting arguments, and streamlines the exposition. The main separation results and numerical findings are unchanged

详情

展开后加载摘要…

URL PDF HTML 收藏