Skip to main content
. Author manuscript; available in PMC: 2022 May 20.
Published in final edited form as: Proc Mach Learn Res. 2022 Mar;151:9304–9333.

Figure 5:

Figure 5:

Results from all folds and parameter choices on real datasets: COMPAS and NETHERLANDS. We can see from the first and second columns (training and test accuracies) that MCP and LASSO do not perform well. Our methods and L0Learn (overlapping) outperform all other methods. Our methods are more computationally efficient than L0Learn.