I will release a package(DHIPF-RNN) in Python AFTER two related papers got published in 2020.
Dr. Gallivan and Dr. BaoAbout me: I am doing research on Nonlinear Filtering Algorithm under the supervision of
My research > A new algorithm (Drift homotopy implicit particle filter or DHIPF) is proposed which is a combination of Implicit sampling and homotopy method. We showed by experiments that DHIPF is faster than drift homotopy particle filters in computations while at same accuracy level. And DHIPF is more accurate (in the measure of Mean square error ) than IPF when we have sparse observations and accurate prediction dynamics. > A new algorithm to improve RNN in stochastic process predictions is proposed. By representing the recurrent information(a deterministic vector) by a distribution which approximated by particles, we could improve the training process of RNN
- A.D., N. De Freitas & N.J. Gordon, SMC Methods in Practice, 2001.
Large collection of chapters on the subject, a bit outdated now but good to start with.
My favorate book. A self-contained survey, which systematically investigate the roots of Bayesian filtering as well as its rich leaves in the literature.
- P. Del Moral, Feynman-Kac Formulae: Genealogical and Interacting Particle Approximations,2004.
Everything you want to know about the theory behind SMC, also includes nice non-standard applications. The notation can appear a little bit overwhelming at the beginning but anyone who has worked on the subject learn to appreciate how powerful they are eventually.
- O. Cappe, E. Moulines & T. Ryden, Inference in Hidden Markov Models, 2005.
A comprehensive treatment of hidden Markov models which includes a few chapters on SMC methods.
- S. Särkkä, Bayesian Filtering and Smoothing, 2013.
An introduction to advanced nonlinear filtering methods including SMC supported by Matlab examples.
A table of researchers and their focus in Data Assimilation
(click author's name to reach to a specific list of papers.)
|Name||Research in Data Assimilation||Group|
|Andrew Stuart||Data Assimilation as Inverse Problem||Caltech|
|Arnaud Doucet; Nando De Freitas; Chris Snyder; Neil Gordon||SMC;MCMC||Cambridge & Oxford|
|Alexandre Chorin; Xuemin Tu; Matthias Morzfield||Implicit Particle Filter||UC-Berkeley|
|Michael Pitt; Neil Shephard||Auxiliary Particle Filter||Harvard|
|Maroulas; Panos Stinis; Feng Bao||Drift Homotopy Particle Filter||UT-Knoxville & ORNL|
|Ramon Van Handel; Xin Tong; Patrick Rebeschini||High Dimensional Particle Filter, Local Method||Princeton|
|Geir Evensen||Ensemble Kalman Filter||NORCE|
|PJ Van Leeuven||Equal Weight, High Dimensional Particle Filter||Univ. of Reading|
|Shijun Liao||Homotopy Method||SJTU|
|Pierre Del Moral||Theoretical Data Assimilation (Feyman-Kac formulae)||INRIA Bordeaux|