You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This repository contains the results of shape-constrained polynomial regression (SCPR) on datasets of our shape-constrained regression benchmark SCR-Benchmark.
Shape-constrained polynomial regression (SCPR)
The hyperparameters of SCPR are the total degree- $d$
of the polynomial, $\lambda$, to determine the strength of regularization, and $\alpha$ to
balance the 1-norm and the 2-norm penalties. We solve the semidefinite programming problem using the commercial solver Mosek. Cf. Hall for a detailed
mathematical description of SCPR.
About
Shape-constrained polynomial regression (SCPR) results on our SCR-Benchmark.