Computational Doob's $h$-transforms for Online Filtering of Discretely Observed Diffusions

This is a PyTorch implementation of the methodology described in the article.

Run the following Python notebooks to repeat experiments for each model:

1. repeat_OU.ipynb and repeat_OU_robust.ipynb for Ornstein-Uhlenbeck model;
2. repeat_logistic.ipynb and repeat_logistic_robust.ipynb for a logistic diffusion model considered in Knape and Valpine (Ecology, 2012);
3. repeat_cell.ipynb and repeat_cell_robust.ipynb for the cell differentiation and development model of Wang et al. (PNAS, 2011).