4 papers · 1 filter
Accelerating Non-Smooth and Heavy-Tailed Sampling
Pervez Ali, Xiaoyu Wang, Yingli Wang +1
Anchored Langevin dynamics (ALD) is useful for non-smooth sampling where the density of the target distribution is possibly non-differentiable and heavy-tailed; reflected anchored…
Improved Analysis for Hessian-free High-resolution Monte Carlo Sampling
Wujun Lv, Xiaoyu Wang, Yingli Wang +1
Hessian-free high-resolution (HFHR) dynamics augments underdamped Langevin dynamics (ULD) with reversible position diffusion for sampling problems that arise in machine learning. W…
Variance Reduction for Stochastic Gradient Generalized Non-reversible Langevin Monte Carlo Algorithms
Bingye Ni, Xiaoyu Wang, Yingli Wang +1
We study the leading-order fluctuation of stochastic gradient Euler-Maruyama estimators for generalized non-reversible Langevin dynamics. Under structural assumptions tailored to t…
Accelerating Constrained Sampling: A Large Deviations Approach
Yingli Wang, Changwei Tu, Xiaoyu Wang +1
The problem of sampling a target probability distribution on a constrained domain arises in many applications including machine learning. For constrained sampling, various Langevin…