首页 > AI前沿 > Finite speed Ornstein-Uhlenbeck-Kac processes and their applications in generative modeling

Finite speed Ornstein-Uhlenbeck-Kac processes and their applications in generative modeling

arXiv机器学习 2026-11-13 08:49 9 阅读 查看原文
用户希望我作为一个专业的新闻排版工程师,将提供的一段纯文本新闻正文排版成语义化的HTML片段。 规则如下: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释。 2. 每个自然段落用`

`包裹,长文用`

`/`

`划分小节。 3. 列表内容用`
  • `,引用用`
    `,代码用`
    `,重要句子可用``。
    4. 保留原文所有信息,不得增删改内容。
    5. 不要输出``/``/``等文档级标签。
    
    正文内容:
    We study the Ornstein-Uhlenbeck-Kac (OU-K) process, where instead of a Brownian motion, a finite-speed Kac process defines the driving process. The OU-K process can be analyzed through a rescaled, time-inhomogeneous Kac process, characterized by a switching rate from the Kac part, and a convergence rate from the Ornstein-Uhlenbeck (OU) part. The law of this inhomogeneous Kac process describes the solution of a wave equation, the Euler-Poisson-Darboux (EPD) equation. For Kac switching rates being multiples of the OU convergence rate, we can explicitly derive this EPD solution via the sum of (the derivative of) two Legendre polynomials, restricted between two Dirac points representing the wave fronts. Using the framework of piecewise-deterministic Markov processes (PDMP), we describe the conditional velocity fields of the OU-K process via our polynomial-law-formula. This allows us to apply the OU-K process in a generative flow model. Since the OU-K velocity fields prove to be bounded by the finite Kac speed, the usual time truncations of diffusion models become unnecessary. We demonstrate advantages over diffusion and flow matching models in terms of generation sharpness on singular data sets, while achieving competitive results on CIFAR-10 image generation.
    
    分析正文:
    这是一段学术/技术新闻(或摘要),包含几个逻辑点:
    1. 介绍OU-K过程及其驱动机制。
    2. 分析方法(重缩放、时间非齐次Kac过程,特征为切换率和收敛率)。
    3. 描述波方程(EPD方程)的解。
    4. 在特定条件下推导EPD解(Legendre多项式)。
    5. 使用PDMP框架描述条件速度场。
    6. 应用于生成流模型,及其优势(无需时间截断,生成锐度优势,CIFAR-10结果)。
    
    因为文本不长,可以划分为几个小节(使用`

    `)或者直接作为几个段落。为了符合“长文用`

    `/`

    `划分小节”的要求,我可以把这段文本适当划分为几个小节,并加上小节标题,但规则4要求“不得增删改内容”。如果我添加了标题,算不算“增”? “不得增删改内容”通常指不改变正文的字词。如果加小标题,可能会被视为增加内容。为了安全起见,不添加额外的标题文字,或者将现有的句子作为段落,必要时用``强调关键术语。不过规则2说“长文用`

    `/`

    `划分小节”,如果不加标题,怎么划分小节? 也许我可以直接把每句话或几个连贯的句子作为`

    `,不强行加标题,因为原文没有标题。或者,我可以把核心概念用``包裹。 “不得增删改内容”意味着不能加小标题,因为小标题是原文没有的。所以最好只用`

    `来包裹自然段落,并对重要句子或术语使用``。 原文是一整段,我可以把它拆分成几个自然段(按照逻辑层次),但不要增加文字。 段落1:We study the Ornstein-Uhlenbeck-Kac (OU-K) process, where instead of a Brownian motion, a finite-speed Kac process defines the driving process. The OU-K process can be analyzed through a rescaled, time-inhomogeneous Kac process, characterized by a switching rate from the Kac part, and a convergence rate from the Ornstein-Uhlenbeck (OU) part. 段落2:The law of this inhomogeneous Kac process describes the solution of a wave equation, the Euler-Poisson-Darboux (EPD) equation. For Kac switching rates being multiples of the OU convergence rate, we can explicitly derive this EPD solution via the sum of (the derivative of) two Legendre polynomials, restricted between two Dirac points representing the wave fronts. 段落3:Using the framework of piecewise-deterministic Markov processes (PDMP), we describe the conditional velocity fields of the OU-K process via our polynomial-law-formula. This allows us to apply the OU-K process in a generative flow model. 段落4:Since the OU-K velocity fields prove to be bounded by the finite Kac speed, the usual time truncations of diffusion models become unnecessary. We demonstrate advantages over diffusion and flow matching models in terms of generation sharpness on singular data sets, while achieving competitive results on CIFAR-10 image generation. 重要句子/术语用``: - Ornstein-Uhlenbeck-Kac (OU-K) process - finite-speed Kac process - switching rate / convergence rate - Euler-Poisson-Darboux (EPD) equation - Legendre polynomials - piecewise-deterministic Markov processes (PDMP) - conditional velocity fields - generative flow model - time truncations 检查规则: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释。 -> 好的。 2. 每个自然段落用`

    `包裹,长文用`

    `/`

    `划分小节。 -> 原文不是长文,也没有现成的小节标题,强行加标题会违反“不得增删改内容”。所以只用`

    `。 3. 列表、引用、代码等这里没有,重要句子用``。 4. 保留原文所有信息,不得增删改内容。 -> 确保文字完全一致,标点也一致。 5. 不要输出文档级标签。 排版后的HTML:

    We study the Ornstein-Uhlenbeck-Kac (OU-K) process, where instead of a Brownian motion, a finite-speed Kac process defines the driving process. The OU-K process can be analyzed through a rescaled, time-inhomogeneous Kac process, characterized by a switching rate from the Kac part, and a convergence rate from the Ornstein-Uhlenbeck (OU) part.

    The law of this inhomogeneous Kac process describes the solution of a wave equation, the Euler-Poisson-Darboux (EPD) equation. For Kac switching rates being multiples of the OU convergence rate, we can explicitly derive this EPD solution via the sum of (the derivative of) two Legendre polynomials, restricted between two Dirac points representing the wave fronts.

    Using the framework of piecewise-deterministic Markov processes (PDMP), we describe the conditional velocity fields of the OU-K process via our polynomial-law-formula. This allows us to apply the OU-K process in a generative flow model.

    Since the OU-K velocity fields prove to be bounded by the finite Kac speed, the usual time truncations of diffusion models become unnecessary. We demonstrate advantages over diffusion and flow matching models in terms of generation sharpness on singular data sets, while achieving competitive results on CIFAR-10 image generation.

    检查文字是否完全和原文一致: 原文:We study the Ornstein-Uhlenbeck-Kac (OU-K) process, where instead of a Brownian motion, a finite-speed Kac process defines the driving process. The OU-K process can be analyzed through a rescaled, time-inhomogeneous Kac process, characterized by a switching rate from the Kac part, and a convergence rate from the Ornstein-Uhlenbeck (OU) part. 一致。 原文:The law of this inhomogeneous Kac process describes the solution of a wave equation, the Euler-Poisson-Darboux (EPD) equation. For Kac switching rates being