SLOPE
SLOPE copied to clipboard
Sorted L1 Penalized Estimation
We should use the duality gap as a stopping criterion instead of the pseudo-gap + infeasibility that we're currently using.
The SLOPE package appears to be slow for some people because they use R's internal BLACK/LAPACK packages, which are far inferior to OPENBLAS. If we switch to Eigen, we can...
`trainSLOPE()` appears to select the wrong optimum in terms of AUC.
Adding the argument 'only_nonzeros' to coef() to print only nonzero coefficients. GSoC 2021 task for the project 'Adding Adaptive Bayesian SLOPE (ABSLOPE) to the SLOPE Package'.
Add argument to `coef()` to print only nonzero coefficients. Mock-up: ```r fit
In SLOPE version 0.3.0 and above, the penalty in the SLOPE objective is scaled depending on the type of scaling that is used in the call to `SLOPE()`. The behavior...
This is relatively easy to do with first-order methods but hard with ADMM.
It is possible that we might want to vary penalty parameter across iterations for the implementation in ADMM. See the excerpt below from the ADMM monograph by Boyd et al....