Plugging O(N) smoothing algos in PGibbs #92
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
…cheap models (as the cost of sampling a single trajectory is still O(N), not O(1)) but useful for models with expensive transition dynamics.
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Outdated suggestions cannot be applied.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.
This makes the new O(N) smoothing algos available for backward sampling in CSMC.
This is not very useful for cheap models, as the cost of sampling a single trajectory is still O(N) (because we need to sample from a categorical of the filtering weights), not O(1) but useful for models with expensive transition dynamics, in which case the cost is dominated by evaluating many possible trajectories.
I'm currently working with such a model in a nested particle filter context, where the cost of evaluating a transition is O(M) where M is the number of inner particles to the PF. This has made a massive difference.