# exact model
f(x) = sum(x.^2)
N = 100
D = 11
X = randn(N, D) # training data
# Method to approximate f
method = KernelInterpolation(f, X)
# test data
X_test = rand(97, D)
# approximated values
ff = evaluate(X_test, method)
jmejia8 / biapprox Goto Github PK
View Code? Open in Web Editor NEWApproximation methods for Bilevel Optimization Algorithms
License: MIT License