Incrementally/gradually updating the embedding as distances matrix changes

I have a use case where I'm passing a pre-computed distance matrix to UMAP.
My distance matrix gradually changes. Is there a way to update the embedding after these incremental changes in distance matrix, without recomputing the whole embedding from scratch? My hope is these incremental updates will be faster than recomputing from scratch.

No files, tests, or entry points are named. Start by examining how UMAP consumes a pre-computed distance matrix and whether an existing embedding can be updated incrementally. Done would require a clearly defined, supported approach for updating the embedding without recomputing it from scratch.

Written by the indexing model from the issue text.