cbg-ethz / bmi

Mutual information estimators and benchmark
https://cbg-ethz.github.io/bmi/
MIT License
26 stars 4 forks source link

Implement a (hidden) Markov model #164

Open pawel-czyz opened 2 weeks ago

pawel-czyz commented 2 weeks ago

Use TFP on JAX to define a new BMM:

X1 – X2 – ... – Xn 
 |   |          |
Y1   Y2 – ... –  Yn

However, this requires first fixing #161, so that the joint distribution returns tuples of arrays for $(X, Y)$, rather than sampling them jointly and unwrapping.