PulseAugur
实时 16:17:50
English(EN) Reliability of Probabilistic Emulation of Physical Systems

新框架AutoCast评估概率预测的可靠性

一个名为AutoCast的新框架已被开发出来,用于评估物理系统概率预测的可靠性。该研究将生成模型(如扩散模型和流匹配模型)与使用CRPS损失训练的确定性模型集成进行比较。结果表明,与在潜在空间中训练的生成模型相比,使用CRPS训练的集成模型通常能提供更可靠的不确定性估计和更快的推理速度。当生成模型在环境空间中训练时,它们显示出相当的覆盖率,但延迟较高。 AI

影响 这项研究提供了一个评估AI驱动的概率预测可靠性的框架,有可能提高其在物理系统建模中的准确性和可信度。

排序理由 该集群包含一篇学术论文,详细介绍了用于物理系统概率模拟的新框架和方法评估。

在 arXiv stat.ML 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

报道来源 [2]

  1. arXiv stat.ML TIER_1 English(EN) · Sam F. Greenbury (The Alan Turing Institute), Radka Jersakova (The Alan Turing Institute), Paolo Conti (The Alan Turing Institute, Autodesk Research), Marjan Famili (The Alan Turing Institute, PhysicsX), Christopher Iliffe Sprague (The Alan Turing Instit… ·

    Reliability of Probabilistic Emulation of Physical Systems

    arXiv:2606.12997v1 Announce Type: cross Abstract: Two dominant approaches have emerged for generating probabilistic forecasts of physical systems: generative models, such as diffusion or flow matching; and ensembles of deterministic models with stochasticity injected, trained usi…

  2. arXiv stat.ML TIER_1 English(EN) · Jason D. McEwen ·

    Reliability of Probabilistic Emulation of Physical Systems

    Two dominant approaches have emerged for generating probabilistic forecasts of physical systems: generative models, such as diffusion or flow matching; and ensembles of deterministic models with stochasticity injected, trained using the continuous ranked probability score (CRPS) …