Skip to content

Estimation, filtering, and inference for partially-observed Markov processes via a two-stage algorithm involving gradient descent (with a novel gradient estimate for the particle filter) warm-started by iterated filtering. Original engine behind the pypomp Python package. Source code for a submission; currently R&R at JRSS-B.

Notifications You must be signed in to change notification settings

hetankevin/diffPomp

Repository files navigation

Code and Manuscript Files for "Automatic Differentiation Accelerates Inference for Partially-Observed Markov Processes

Thank you for reviewing the code submitted with the manuscript. The LaTeX source files for the manuscript are included in the paper/ directory. The source code for the simulations are contained in filtering.py, optim.py, pomps.py, and resampling.py, along with the three driver Jupyter notebooks used for generating the results in cholera.ipynb, cholera_biasvar.ipynb, and cholera_npeb.ipynb.

Please contact the authors if you have any further questions.

About

Estimation, filtering, and inference for partially-observed Markov processes via a two-stage algorithm involving gradient descent (with a novel gradient estimate for the particle filter) warm-started by iterated filtering. Original engine behind the pypomp Python package. Source code for a submission; currently R&R at JRSS-B.

Topics

Resources

Stars

Watchers

Forks

Packages

No packages published

Contributors 2

  •  
  •