Constrained or partitioned inference for Turing models.

TuringLang / Turing.jl

Bayesian inference with probabilistic programming.

https://turinglang.org

MIT License

2.03k stars 218 forks source link

Constrained or partitioned inference for Turing models. #2249

Open yebai opened 3 months ago

yebai commented 3 months ago

We currently have options to allow users to specify constraints for MAP/MLE optimisation-based learning.

https://github.com/TuringLang/docs/pull/470

It would be interesting to also allow users to specify constraints for MCMC sampling, e.g., each chain is confined to sampling from a partition. Then, the user can merge MCMC samples from different partitions somehow.