用于变分序列蒙特卡洛的可微系统重采样
Differentiable Systematic Resampling for Variational Sequential Monte Carlo
浏览论文内容
中文总结 AI 辅助
针对粒子滤波器重采样步骤离散阻碍变分序列蒙特卡洛梯度学习的问题，提出可微系统重采样DSR，其计算开销低且性能相当或更优，可用于滤波与动力学学习。
中文摘要 AI 辅助
粒子滤波器是非线性状态估计的标准工具，但其重采样步骤是离散的，阻碍了变分序列蒙特卡洛中基于梯度的学习。我们提出可微系统重采样（Differentiable Systematic Resampling，DSR），一种受温度控制的系统重采样松弛方法，它保留了系统重采样的CDF有序带状结构，同时实现了完整的梯度流。当温度趋近于零时，DSR收敛到精确的系统重采样，且我们证明了其诱导偏差的逐点指数收敛速率。与基于最优传输的可微重采样相比，DSR避免了迭代求解器，计算开销显著更低。在随机动力系统和真实世界手写数据上的实验表明，DSR实现了相当或更优的滤波与动力学学习性能。
英文摘要
Particle filters are a standard tool for nonlinear state estimation, but their resampling step is discrete, preventing gradient-based learning in variational sequential Monte Carlo. We introduce Differentiable Systematic Resampling (DSR), a temperature-controlled relaxation of systematic resampling, that preserves the CDF-ordered, banded structure of systematic resampling while enabling full gradient flow. DSR converges to exact systematic resampling as the temperature vanishes, and we prove a pointwise exponential convergence rate for the induced bias. Compared to optimal-transport-based differentiable resampling, DSR avoids iterative solvers and has substantially lower computational overhead. Experiments on stochastic dynamical systems and real-world handwriting data show that DSR achieves comparable or superior filtering and dynamics learning performance.
发表机构
- KTH Royal Institute of Technology（瑞典皇家理工学院）
机构由 AI 辅助整理，请以论文原文为准。
