arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.14780cs.LGstat.ML
收藏

度量流匹配用于数据流形上的平滑插值

Metric Flow Matching for Smooth Interpolations on the Data Manifold

University of Oxford(牛津大学);Mila(米拉(魁北克人工智能研究所));Université de Montréal(蒙特利尔大学);AITHYRA(艾西拉(机构名))

Kacper Kapuśniak, Peter Potaptchik, Teodora Reu, Leo Zhang, Alexander Tong, Michael Bronstein, Avishek Joey Bose, Francesco Di Giovanni

AI总结:

本文提出度量流匹配(MFM),一种无模拟的条件流匹配框架,通过最小化数据诱导黎曼度量的动能学习近似测地线插值,使生成模型匹配数据流形上的向量场,在LiDAR导航、非配对图像转换和细胞动力学建模等任务中优于欧几里得基线,并在单细胞轨迹预测上达到最先进性能。

摘要:

匹配目标函数是现代生成模型成功的基础,并依赖于构建将源分布转换为目标分布的条件路径。尽管条件路径是基本构建模块,但其设计主要基于欧几里得几何假设,导致产生直线插值。然而,这对于轨迹推断等任务可能特别受限,因为直线路径可能位于数据流形之外,从而无法捕捉产生观测边缘分布的潜在动态。在本文中,我们提出了度量流匹配(Metric Flow Matching, MFM),一种新颖的无模拟条件流匹配框架,其中插值是通过最小化数据诱导的黎曼度量的动能来学习的近似测地线。这样,生成模型匹配数据流形上的向量场,这对应于更低的不确定性和更有意义的插值。我们规定了与任务无关的通用度量来实例化MFM,并在包括LiDAR导航、非配对图像转换和细胞动力学建模在内的一系列具有挑战性的问题上进行了测试。我们观察到MFM优于欧几里得基线,特别是在单细胞轨迹预测上达到了最先进的性能。

英文摘要:

Matching objectives underpin the success of modern generative models and rely on constructing conditional paths that transform a source distribution into a target distribution. Despite being a fundamental building block, conditional paths have been designed principally under the assumption of Euclidean geometry, resulting in straight interpolations. However, this can be particularly restrictive for tasks such as trajectory inference, where straight paths might lie outside the data manifold, thus failing to capture the underlying dynamics giving rise to the observed marginals. In this paper, we propose Metric Flow Matching (MFM), a novel simulation-free framework for conditional flow matching where interpolants are approximate geodesics learned by minimizing the kinetic energy of a data-induced Riemannian metric. This way, the generative model matches vector fields on the data manifold, which corresponds to lower uncertainty and more meaningful interpolations. We prescribe general metrics to instantiate MFM, independent of the task, and test it on a suite of challenging problems including LiDAR navigation, unpaired image translation, and modeling cellular dynamics. We observe that MFM outperforms the Euclidean baselines, particularly achieving SOTA on single-cell trajectory prediction.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑