Fix Dead Mode Removal #1999
Merged
Fix Dead Mode Removal #1999
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…dding the marginals for future factors
Merged
dellaert
reviewed
Jan 25, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Currently, I am encountering an issue where if we remove a dead mode and then we add a factor which has the mode key in it, the elimination algorithm errors out in the construction of the elimination tree. This is particularly true if we add a discrete chain, where we have conditionals like$P(m_{k+1} | m_k)$ and $m_k$ was the dead mode removed.
The simplest fix is to add the mode marginals to the
HybridBayesNetwhen we remove the mode from the joint so that any future factors that get added referencing those modes don't complain during elimination.I also added a check for the joint discrete conditional being empty (aka no keys) and we no longer add it. A new unit test was added to confirm this works, but failed before.