Last updated: 2018-07-27

workflowr checks: (Click a bullet for more information)
Expand here to see past versions:


In Progress

Investigation 5a. We should change the default init_fn to udv_si_svd when there is missing data and udv_svd otherwise.

Investigation 5b. More evidence supporting a change to the default initialization function.

Investigation 6. Poor optim results can produce large decreases in the objective function. We should use warmstarts when ebnm_fn = ebnm_pn.

Investigation 7. The advantages of warmstarts are not nearly as compelling when ebnm_fn = ebnm_ash.

Still Relevant

Note 1. Notes on computing the FLASH objective function. I derive an explicit expression for the KL divergence between prior and posterior.

Note 2. An alternate algorithm for optimizing the FLASH objective, using the explicit expression derived in the previous note.

Investigation 4. The alternate algorithm agrees with FLASH with respect to both the objective and fit obtained.

Archived

The bug causing the problem described in Investigations 1-3 was fixed in version 0.1-13 of package ebnm.

Investigation 1. The FLASH objective function can behave very erratically.

Investigation 2. This problem only occurs when using ebnm_pn, not ebnm_ash.

Investigation 3. The objective can continue to get worse as loadings are repeatedly updated. Nonetheless, convergence takes place (from above!).


This reproducible R Markdown analysis was created with workflowr 1.0.1