You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The query() method for the InferenceEngine currently supports the computation of the marginal likelihood of states in the network given observations joined by conjunctions (ANDs). Support for disjunctive observations would allow for greater insights to be gained from the network.
Context
This extension would allow for more functionality within CausalNex for examining different scenarios.
The text was updated successfully, but these errors were encountered:
Hi @NShah19, thanks for your suggestion - this is certainly an interesting point. To our best knowledge, however, the standard notations for conditional/joint probabilities and Bayesian inference in the literature are defined based on conjunctive operation.
Would you be able to perhaps suggest a reference paper and/or use case example for us to investigate further? It might be possible to make use of the inclusion-exclusion principle to do this, but we would like to make sure that the inference mechanism is general enough if we were to implement it.
Description
The query() method for the InferenceEngine currently supports the computation of the marginal likelihood of states in the network given observations joined by conjunctions (ANDs). Support for disjunctive observations would allow for greater insights to be gained from the network.
Context
This extension would allow for more functionality within CausalNex for examining different scenarios.
The text was updated successfully, but these errors were encountered: