arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2410.09246cs.LGcs.AIstat.ML
收藏

DFM:无插值的对偶流匹配

DFM: Interpolant-free Dual Flow Matching

Panasonic AI Lab(松下人工智能实验室);Panasonic Holdings Corporation(松下控股公司)

Denis Gudovskiy, Tomoyuki Okuno, Yohei Nakata

AI总结:

针对连续正则化流求解ODE计算代价高且依赖插值的问题,提出无插值的对偶流匹配方法,通过联合优化前向与反向向量场促进双射性,在SMAP无监督异常检测中取得SOTA性能。

摘要:

连续正则化流(CNF)能够用富有表现力的无限长度架构对数据分布进行建模。但这种建模在最大似然训练期间涉及计算代价高昂的求解常微分方程(ODE)过程。最近提出的流匹配(FM)框架允许利用带有插值前向向量场的回归目标来大幅简化训练阶段。在本文中,我们提出了一种无插值的对偶流匹配(DFM)方法,无需对所建模的向量场进行显式假设。DFM优化前向以及额外的反向向量场模型,使用的新目标促进了前向与反向变换的双射性。我们在SMAP无监督异常检测上的实验表明,与使用最大似然或FM目标训练的CNF相比,DFM具有优势,并取得了最先进的性能指标。

英文摘要:

Continuous normalizing flows (CNFs) can model data distributions with expressive infinite-length architectures. But this modeling involves computationally expensive process of solving an ordinary differential equation (ODE) during maximum likelihood training. Recently proposed flow matching (FM) framework allows to substantially simplify the training phase using a regression objective with the interpolated forward vector field. In this paper, we propose an interpolant-free dual flow matching (DFM) approach without explicit assumptions about the modeled vector field. DFM optimizes the forward and, additionally, a reverse vector field model using a novel objective that facilitates bijectivity of the forward and reverse transformations. Our experiments with the SMAP unsupervised anomaly detection show advantages of DFM when compared to the CNF trained with either maximum likelihood or FM objectives with the state-of-the-art performance metrics.

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。

↑