This paper works on a heuristic algorithm with determinacy for the global optimization of Continuous PieceWise Linear (CPWL) programming. The widely applied CPWL programming can be equivalently transformed into D.C. programming and concave optimization over a polyhedron. Considering that the super-level sets of concave piecewise linear functions are polyhedra, we propose the Hill Tunneling via Weighted Simplex Centroid (HTWSC) algorithm, which can escape a local optimum to reach the other side of its contour surface by cutting across the super-level set. The searching path for hill tunneling is established via the weighted centroid of a constructed simplex. In the numerical experiments, different weighting methods are studied first, and the best is chosen for the proposed HTWSC algorithm. Then, the HTWSC algorithm is compared with the hill detouring method and the software CPLEX for the equivalent mixed integer programming, with results indicating its superior performance in terms of numerical efficiency and the global search capability.
- Article type
- Year
- Co-author
In order to recover a signal from its compressive measurements, the compressed sensing theory seeks the sparsest signal that agrees with the measurements, which is actually an
This paper works on a modified simplex algorithm for the local optimization of Continuous PieceWise Linear (CPWL) programming with generalization of hinging hyperplane objective and linear constraints. CPWL programming is popular since it can be equivalently transformed into difference of convex functions programming or concave optimization. Inspired by the concavity of the concave CPWL functions, we propose an Objective Variation Simplex Algorithm (OVSA), which is able to find a local optimum in a reasonable time. Computational results are presented for further insights into the performance of the OVSA compared with two other algorithms on random test problems.
Memristor is introduced as the fourth basic circuit element. Memristor exhibits great potential for numerous applications, such as emulating synapse, while the mathematical model of the memristor is still an open subject. In the linear-drift model, the boundary condition of the device is not considered. This paper proposes an extended linear-drift model of the memristor. The extended linear-drift model keeps the linear characteristic and simplicity of the linear-drift model and considers the boundary condition of the device. A piecewise linear approximation model of the extended linear-drift model is given. Both models are suitable for describing the memristor.