Dynamic Time Warping (DTW) is a common choice for measuring the discrepancy between time series. It proceeds by finding an optimal warping between timestamps through dynamic programming, which can be slow and is not differentiable. Consequently, several variants have been proposed. Among them, Soft-DTW proposes a smooth formulation of DTW, amenable to gradient computation, while Continuous-DTW (CDTW) introduces a continuous formulation through parametrizable warping functions. Yet, computational costs remain high for both methods and impractical for large datasets in modern machine learning pipelines. This paper revisits CDTW in several ways. After establishing its differentiability under mild conditions, we introduce FastCDTW, which is an efficient approximate solver with two different computational strategies: an exact closed-form scheme for piecewise-linear warping functions, and an unbiased Monte Carlo estimator for smooth ones, such as implicit neural networks. Both strategies provide gradients with respect to the warping function and the input time series, and the resulting algorithm is GPU-friendly. We demonstrate its benefits through its integration into more complex learning problems, including time series averaging, classification and deep domain adaptation.
Learning non-Gaussian Time Series using the Box-Cox Gaussian Process
Center for Mathematical Modeling and Department of Mathematical Engineering Universidad de Chile(智利大学数学建模中心和数学工程系);Center for Mathematical Modeling Universidad de Chile(智利大学数学建模中心)
Fast Efficient Hyperparameter Tuning for Policy Gradients
University of Oxford(牛津大学)
策略梯度方法的性能对必须针对任何新应用程序进行调优的超参数设置很敏感。用于调整超参数的广泛使用的网格搜索方法是样本效率低且计算昂贵的。更先进的方法,如基于人口的训练,学习超参数的最佳时间表,而不是固定的设置,可以产生更好的结果,但也是样本效率低,计算成本高。在本文中,我们提出了Hyperparameter Optimization on the Fly(HOOF),这是一种无梯度算法,它只需要一次训练就可以自动调整直接通过梯度影响策略更新的超参数。其主要思想是使用现有的轨迹采样的策略梯度方法来优化一步改进目标,产生一个样本和计算效率高的算法,易于实现。我们在多个领域和算法上的实验结果表明,使用HOOF来学习这些超参数计划可以提高学习速度和性能。
Stochastic Gradient Descent: Going As Fast As Possible But Not Faster
UCSF(加州大学旧金山分校);INRIA-CNRS-UPSud-UPSay(法国国家信息与自动化研究所-法国国家科学研究中心-巴黎南大学-巴黎萨克雷大学);TAU, U. Paris-Sud(计算机科学研究实验室,巴黎南大学);CNRS-UPSud-INRIA-UPSay(法国国家科学研究中心-巴黎南大学-法国国家信息与自动化研究所-巴黎萨克雷大学)