Two novel schemes of selecting the current best solutions for multiobjective differential evolution are proposed in this paper. Based on the search biases strategy suggested by Runarsson and Yao, a hybrid of multiobjective differential evolution and genetic algorithm with (N+N) framework for constrained MOPs is given. And then the hybrid algorithm adopting the two schemes respectively is compar...

The aim of this paper is to generalize the approaches used for developing hybrid intelligent systems based on integration of neural networks, fuzzy logic and genetic algorithms. The paper introduces the concepts of intelligent artificial life system (IALS) space as a generalized conceptual space which represents a framework for all possible integration schemes of such intelligence technologies....

Model-based diagnosis schemes for hybrid systems are inherently complex because they deal with models that exhibit continuous and discrete dynamics. Our approach to hybrid system diagnosis combines qualitative hypothesis refinement and quantitative parameter estimation techniques. The goal is to implement online schemes that are computationally feasible. We have developed an approach to improve...

SUMMARY Numerical Stability of the Finite Element/Finite Difference Time Domain Hybrid algorithm is dependent on the hybridization mechanism adopted. A framework is developed to analyze the numerical stability of the hybrid time marching algorithm. First, the global iteration matrix representing the hybrid algorithm following different hybridization schemes is constructed. An analysis of the ei...

A numerical investigation of high-resolution schemes for solving the compressible Navier-Stokes equations in the context of Implicit Large Eddy Simulation (ILES) is presented. Computations have been performed using a hybrid total variation-diminishing (TVD) scheme. The hybrid TVD scheme combines third-order Godunov-type fluxes with first-order (dissipative) schemes using the flux limiters appro...

In this paper we examine the notion of plaintext awareness as it applies to hybrid encryption schemes. We apply this theory to the Cramer-Shoup hybrid scheme acting on fixed length messages and deduce that the Cramer-Shoup scheme is plaintext-aware in the standard model. This answers a previously open conjecture of Bellare and Palacio on the existence of fully plaintext-aware encryption schemes.

For finite geometry low-density parity-check codes, heavy row and column weights in their parity check matrix make the decoding with even Min-Sum (MS) variants computationally expensive. To alleviate it, we present a class of hybrid schemes by concatenating a parallel bit flipping (BF) variant with an Min-Sum (MS) variant. In most SNR region of interest, without compromising performance or conv...

Tseng's forward-backward-forward splitting method for finding zeros of the sum Lipschitz continuous monotone and maximal operators is known to converge weakly in infinite dimensional Hilbert spaces. The inertial viscosity approximation techniques are widely used accelerate iterative algorithms obtain strong convergence, respectively. In this paper, we propose two fast, strongly convergent modif...