Differentiable Particle Filtering via Entropy-Regularized Optimal Transport

02/15/2021
by   Adrien Corenflos, et al.
7

Particle Filtering (PF) methods are an established class of procedures for performing inference in non-linear state-space models. Resampling is a key ingredient of PF, necessary to obtain low variance likelihood and states estimates. However, traditional resampling methods result in PF-based loss functions being non-differentiable with respect to model and PF parameters. In a variational inference context, resampling also yields high variance gradient estimates of the PF-based evidence lower bound. By leveraging optimal transport ideas, we introduce a principled differentiable particle filter and provide convergence results. We demonstrate this novel method on a variety of applications.

READ FULL TEXT
research
11/07/2018

Wasserstein Variational Gradient Descent: From Semi-Discrete Optimal Transport to Ensemble Variational Inference

Particle-based variational inference offers a flexible way of approximat...
research
03/29/2021

Variational Rejection Particle Filtering

We present a variational inference (VI) framework that unifies and lever...
research
12/14/2022

Particle-Based Score Estimation for State Space Model Learning in Autonomous Driving

Multi-object state estimation is a fundamental problem for robotic appli...
research
06/03/2019

A scalable optimal-transport based local particle filter

Filtering in spatially-extended dynamical systems is a challenging probl...
research
06/18/2021

Differentiable Particle Filtering without Modifying the Forward Pass

In recent years particle filters have being used as components in system...
research
02/20/2023

Differentiable Bootstrap Particle Filters for Regime-Switching Models

Differentiable particle filters are an emerging class of particle filter...
research
07/03/2023

Transport, Variational Inference and Diffusions: with Applications to Annealed Flows and Schrödinger Bridges

This paper explores the connections between optimal transport and variat...

Please sign up or login with your details

Forgot password? Click here to reset