TY - GEN
T1 - UNROLLING PARTICLES
T2 - 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
AU - Gama, Fernando
AU - Zilberstein, Nicolas
AU - Baraniuk, Richard G.
AU - Segarra, Santiago
N1 - Funding Information:
This work was partially supported by NSF under award CCF-2008555. Email: {fgama,nzilberstein,richb,segarra}@rice.edu
Publisher Copyright:
© 2022 IEEE
PY - 2022
Y1 - 2022
N2 - Particle filtering is used to compute nonlinear estimates of complex systems. It samples trajectories from a chosen distribution and computes the estimate as a weighted average of them. Easy-to-sample distributions often lead to degenerate samples where only one trajectory carries all the weight, negatively affecting the resulting performance of the estimate. While much research has been done on the design of appropriate sampling distributions that would lead to controlled degeneracy, in this paper our objective is to learn sampling distributions. Leveraging the framework of algorithm unrolling, we model the sampling distribution as a multivariate normal, and we use neural networks to learn both the mean and the covariance. We carry out unsupervised training of the model to minimize weight degeneracy, relying only on the observed measurements of the system. We show in simulations that the resulting particle filter yields good estimates in a wide range of scenarios.
AB - Particle filtering is used to compute nonlinear estimates of complex systems. It samples trajectories from a chosen distribution and computes the estimate as a weighted average of them. Easy-to-sample distributions often lead to degenerate samples where only one trajectory carries all the weight, negatively affecting the resulting performance of the estimate. While much research has been done on the design of appropriate sampling distributions that would lead to controlled degeneracy, in this paper our objective is to learn sampling distributions. Leveraging the framework of algorithm unrolling, we model the sampling distribution as a multivariate normal, and we use neural networks to learn both the mean and the covariance. We carry out unsupervised training of the model to minimize weight degeneracy, relying only on the observed measurements of the system. We show in simulations that the resulting particle filter yields good estimates in a wide range of scenarios.
KW - algorithm unrolling
KW - particle filtering
KW - unsupervised learning
UR - http://www.scopus.com/inward/record.url?scp=85131261466&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85131261466&partnerID=8YFLogxK
U2 - 10.1109/ICASSP43922.2022.9747290
DO - 10.1109/ICASSP43922.2022.9747290
M3 - Conference contribution
AN - SCOPUS:85131261466
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 5498
EP - 5502
BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 23 May 2022 through 27 May 2022
ER -