In contrast to their seemingly simple and shared structure of independence and stationarity, Lévy processes exhibit a wide variety of behaviors, from the self-similar Wiener process to piecewise-constant compound Poisson processes. Inspired by the recent paper of Ghourchian, Amini, and Gohari (2018), we characterize their compressibility by studying the entropy of their double discretization (both in time and amplitude) in the regime of vanishing discretization steps. For a Lévy process with absolutely continuous marginals, this reduces to understanding the asymptotics of the differential entropy of its marginals at small times, for which we obtain a new local central limit theorem. We generalize known results for stable processes to the non-stable case, with a special focus on Lévy processes that are locally self-similar, and conceptualize a new compressibility hierarchy of Lévy processes, captured by their Blumenthal-Getoor index.
Time reversal of Markov processes with jumps under a finite entropy condition
基于熵最优输运理论,研究了$\mathbb{R}^n$中Markov跳过程的时间反演问题.借助于Cattiaux,Gentil等人最近得到的马尔可夫过程的carré du champ的抽象分部积分法公式,并利用Föllmer在80年代发现的熵改进策略,计算了一类具有可能无界变化样本路径和奇异跳强度的跳过程的时反过程的半鞅特征.
Deep neural networks with dependent weights: Gaussian Process mixture limit, heavy tails, sparsity and compressibility
KAIST(韩国科学技术院);Huawei Technologies(华为技术有限公司);Fordham University(福特汉姆大学);Kim Jaechul Graduate School of AI, KAIST(韩国科学技术院金在哲人工智能研究生院);Institute for Basic Science (IBS)(基础科学研究院);University of Oxford(牛津大学)
Time reversal of diffusion processes under a finite entropy condition
受熵最优输运理论的启发,本文重新讨论了扩散过程的时间反演。本文导出了抽象空间中马尔可夫过程的carré du champ的分部积分法公式。它导致了一个广泛的一类扩散过程的时间反演公式在$ \mathbb{R}^n$可能与奇异漂移,在这一领域的推广已经知道的结果。分部积分法公式的证明依赖于随机导数。然后,应用该公式计算扩散测度$P$的时间反演$P^*$的半鞅特征,条件是$P$相对于另一扩散测度$R$的相对熵是有限的,并且时间反演$R^*$的半鞅特征是已知的(例如当参考路径测度$R$是可逆的时)。为了说明这种方法的鲁棒性,我们还利用分部积分法公式导出了图上随机游动的时间反演公式。