Variational Formulation of Particle Flow
Yinzhuang Yi [email protected], Jorge Cortés [email protected], Nikolay Atanasov [email protected] · 2025
Read Variational Formulation of Particle Flow by Yinzhuang Yi [email protected], Jorge Cortés [email protected], Nikolay Atanasov [email protected] (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-04007.txt.
Description
Three engineers recast a tracking method called particle flow in the language of variational inference, in a 2025 paper. Particle flows move a swarm of guesses toward the true state of a system, following a path between probability curves. The authors show that this path matches a known gradient flow that steadily lowers the gap between an approximate curve and the true one. That link grounds the method in a wider theory of estimation.
When the approximating curve is a simple bell shape, the flow simplifies into a clean update rule. Under tidy assumptions, it reduces exactly to a well-known filter used in tracking. A mixture of bells extends the method to cases with several plausible states. Tests on estimation problems in low and high dimensions backed the theory. The work gives practitioners a firmer base for a widely used tool.
Catalog record
| CREATOR | Yinzhuang Yi [email protected], Jorge Cortés [email protected], Nikolay Atanasov [email protected] |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 163 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-04007