Bilevel LP-LP (KKT Reformulation)
Problem Definition
This example is taken from [8] and is an LP-LP bilevel problem.
The problem is formulated as follows:
\[\begin{split}\begin{align}
\min_{x, y} \ & x + 6 y \\
\text{s.t.} \ & -x + 5y \le 12.5 \\
& x \ge 0 \\
& y\in
\begin{array}[t]{l}
\displaystyle \underset{y}{\text{arg min}} \ & -y \\
\text{s.t.} \ & 2 x - y \ge 0, \\
& -x - y \ge -6, \\
& -x + 6 y \ge -3, \\
& x + 3 y \ge 3.
\end{array}
\end{align}\end{split}\]
KKT Reformulation with idol
In this example, we show how to model this LP-LP bilevel problem and how to derive its KKT reformulation.
//
// Created by henri on 30.08.24.
//
#include <iostream>
#include <Research/idol/lib/include/idol/modeling.h>
#include "idol/optimizers/bilevel-optimization/wrappers/MibS/MibS.h"
#include "idol/modeling/bilevel-optimization/LowerLevelDescription.h"
#include "idol/optimizers/mixed-integer-optimization/wrappers/Gurobi/Gurobi.h"
#include "idol/modeling/bilevel-optimization/read_from_file.h"
#include "idol/modeling/models/KKT.h"
using namespace idol;
int main(int t_argc, const char** t_argv) {
/*
This example is taken from Kleinert, T. (2021). “Algorithms for Mixed-Integer Bilevel Problems with Convex Followers.” PhD thesis.
min x + 6 y
s.t.
-x + 5 y <= 12.5,
y in argmin { -y :
2 x - y >= 0,
-x - y >= -6,
-x + 6 y >= -3,
x + 3 y >= 3.
}
x >= 0.
*/
Env env;
// Define High Point Relaxation
Model high_point_relaxation(env);
auto x = high_point_relaxation.add_var(0, Inf, Continuous, "x");
auto y = high_point_relaxation.add_var(-Inf, Inf, Continuous, "y");
high_point_relaxation.set_obj_expr(x + 6 * y);
auto follower_c1 = high_point_relaxation.add_ctr(2 * x - y >= 0, "f1");
auto follower_c2 = high_point_relaxation.add_ctr(-x - y >= -6, "f2");
auto follower_c3 = high_point_relaxation.add_ctr(-x + 6 * y >= -3, "f3");
auto follower_c4 = high_point_relaxation.add_ctr(x + 3 * y >= 3, "f4");
high_point_relaxation.add_ctr(-x + 5 * y <= 12.5);
// Prepare bilevel description
Bilevel::LowerLevelDescription description(env);
description.set_follower_obj_expr(- y);
description.make_follower_var(y);
description.make_follower_ctr(follower_c1);
description.make_follower_ctr(follower_c2);
description.make_follower_ctr(follower_c3);
description.make_follower_ctr(follower_c4);
Reformulators::KKT reformulator(high_point_relaxation, description);
Model single_level(env);
reformulator.add_kkt_reformulation(single_level);
std::cout << high_point_relaxation << std::endl;
std::cout << single_level << std::endl;
single_level.use(Gurobi());
single_level.optimize();
std::cout << save_primal(single_level) << std::endl;
return 0;
}