Tighten alpha bound for biclustering
Opened this issue · 1 comments
dansenglund commented
The current alpha bound works in practice, but it is not as tight as theoretically possible. However, tighter bounds seemed to result in periodic failure to converge. Need to determine the right alpha get performance guarantees.
michaelweylandt commented
Let's revisit this as part of adding missing data support, since the necessary alpha
will depend on the missing data pattern as well.