This paper presents a novel multiobjective optimization strategy based on the cross entropy method (MOCE). The cross-entropy (CE) is stochastic learning algorithm inspired from rare event simulations and proved to be successful in solution of difficult single objective real-valued problems. presented work extends use optimization. For this purpose, parameters CE search are adapted using information collected clustered nondominated solutions Pareto front. Comparison with well known algorithms...

Probability collectives (PC) is a recent agent-based search framework for function optimisation through optimising parameters of collection probability distributions. Differential evolution (DE) successful metaheuristic method particularly real-parameter global optimisation. This paper presents hybrid computational model based on modified PC and DE algorithms the purpose improved solutions real-valued problems. In proposed model, performs first phase local explores promising areas updating...

A genetic algorithm employing multiple crossover operators in its implementation is presented. set of available established initially and a particular operator probabilistically selected from this for recombination. Each assigned fitness value based on the amount improvements they achieve over number previous generations. Hence, randomized dynamic selection scheme followed. This approach used solution which optimal results are achieved reasonable computation times even very difficult problem...

A new architecture of deep neural networks, directed acyclic graph convolutional networks (DAG-CNNs), is used to classify heartbeats from electrocardiogram (ECG) signals into different subject-based classes. DAG-CNNs not only fuse the feature extraction and classification stages ECG a single automated learning procedure, but also utilized multi-scale features perform score-level fusion multiple classifiers automatically. Therefore, DAG-CNN negates necessity extract hand-crafted features. In...

An artificial immune system using the clonal selection principle with multiple hypermutation operators in its implementation is presented. Mutation to be used are identified initially. In every mutation operation, fitness gain achieved by employed operator computed and stored. Accordingly, assigned values based on improvements they achieve over a number of previous generations. These determine probabilities. This approach for solution well-known numerical optimization problem, frequency...

A particle swarm optimization strategy using an external memory of partial position and velocity vectors for the purpose achieving better faster search capabilities is introduced. Partially complete stored in are segments cut from two components promising solutions over a number previous iterations, where size location selected completely at random. Elements (segments) also associated with their parents' fitness values that used retrieving elements. After every iteration, worst k% population...

A novel memory-based particle swarm optimization algorithm employing externally implemented global (shared) and particle-based (local) memories a colonization approach similar to artificial immune system algorithms is presented. At any iteration, keep number of previously best performing personal positions for each the memory keeps globally found so far. set velocities computed using within its local randomly selected from memory. This way, colony new obtained one with fitness put swarm....

A new two-layer switchbox/channel router, SAR, based on the simulated annealing algorithm is presented. The routing region represented by a matrix, and nets are handled using dynamic linked list whose nodes either net segments or vias. Search in configuration space performed reshaping layouts of randomly selected making use Lee-type moves with adaptive parameters, that change layer assignments segments. SAR works no user intervention. It completely automatic has problem dependence. For all...

Multiobjective evolutionary optimization has been demonstrated to be an efficient method for some difficult multiobjective problems; particularly the quadratic assignment problem which is a provably NP-complete with multitude of real-world applications. This paper introduces use segment-based external memory in optimization. In principle, variable-size solution segments taken from number previously promising solutions are stored whose elements used construction new solutions. solution,...

provably successful evolutionary optimization tools in the of a Fuzzy-Rule-Base (FRB) for three well known fuzzy modeling inference methods: Zadeh's (center-of-gravity), Kosko's (Standard- Additive-Model), and Takagi-Sugeno's (local linear) Model. In Fuzzy System Modeling an uncertain data, FRB keeps model information within rules. The initial fuzzy-rule-base algorithms is extracted using Bezdek's FCM. optimization, normalized root mean square error training data minimized fine-tuning...

An ant colony optimization approach towards the development of robust game strategies for iterated prisoner's dilemma (IPD) is presented. It demonstrated that, ants can play IPD and, in spite comparably longer running times, provide better quality than genetic algorithms. The success developed strategies, compared to ones using algorithms are illustrated by experimental evaluations.

In this paper an evolutionary multi-objective optimization approach is applied to extend fuzzy multi objective problem considered by Jeng-Jong Lin [1] and Gupta et al [2, 3]. our extended work, the concept of [6, 8] due time increase satisfaction grade customers meet their demands much better as well decrease delay times vehicles, which service during desired period times, so objectives are follows: maximize loading capacity minimize distance travelled waiting customers, vehicles.