• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.9
• /
• pp.2087-2093
• /
• 2014

In this paper, an orthogonal matching pursuit (OMP) method combined with genetic algorithm (GA), named GAOMP, is proposed for sparse signal recovery. Some recent greedy algorithms such as SP, CoSaMP, and gOMP improved the reconstruction performance by deleting unsuitable atoms at each iteration. However they still often fail to converge to the solution because the support set could not avoid the local minimum during the iterations. Mutating the candidate support set chosen by the OMP algorithm, GAOMP is able to escape from the local minimum and hence recovers the sparse signal. Experimental results show that GAOMP outperforms several OMP based algorithms and the $l_1$ optimization method in terms of exact reconstruction probability.

• The Korean Journal of Applied Statistics
• /
• v.26 no.5
• /
• pp.847-856
• /
• 2013

We propose a modified learning process for generalized neural network using a learning algorithm by Liu et al. (2001). We consider the effect of initial weights, training results and learning errors using a modified learning process. We employ an incremental training procedure where training patterns are learned systematically. Our algorithm starts with a single training pattern and a single hidden layer neuron. During the course of neural network training, we try to escape from the local minimum by using a weight scaling technique. We allow the network to grow by adding a hidden layer neuron only after several consecutive failed attempts to escape from a local minimum. Our optimization procedure tends to make the network reach the error tolerance with no or little training after the addition of a hidden layer neuron. Simulation results with suitable initial weights indicate that the present constructive algorithm can obtain neural networks very close to minimal structures and that convergence to a solution in neural network training can be guaranteed. We tested these algorithms extensively with small training sets.

• Journal of the Korea Convergence Society
• /
• v.9 no.2
• /
• pp.13-18
• /
• 2018

This paper is based on a simple approach that the most efficient learning of a multi-layered network is the process of finding the optimal set of weight vectors. To overcome the disadvantages of general learning problems, the proposed model uses a combination of features of EBP and OVSSA. In other words, the proposed method can construct a single model by taking advantage of each algorithm so that it can escape to the probability theory of OVSSA in order to reinforce the property that EBP falls into local minimum value. In the proposed algorithm, methods for reducing errors in EBP are used as energy functions and the energy is minimized to OVSSA. A simple experimental result confirms that two algorithms with different properties can be combined.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.4
• /
• pp.2038-2057
• /
• 2017

Gaussian learning is a new technology in the computational intelligence area. However, this technology weakens the learning ability of a particle swarm and achieves a lack of diversity. Thus, this paper proposes a vector Gaussian learning strategy and presents an effective approach, named particle swarm optimization based on vector Gaussian learning. The experiments show that the algorithm is more close to the optimal solution and the better search efficiency after we use vector Gaussian learning strategy. The strategy adopts vector Gaussian learning to generate the Gaussian solution of a swarm's optimal location, increases the learning ability of the swarm's optimal location, and maintains the diversity of the swarm. The method divides the states into normal and premature states by analyzing the state threshold of the swarm. If the swarm is in the premature category, the algorithm adopts an inertia weight strategy that decreases linearly in addition to vector Gaussian learning; otherwise, it uses a fixed inertia weight strategy. Experiments are conducted on eight well-known benchmark functions to verify the performance of the new approach. The results demonstrate promising performance of the new method in terms of convergence velocity and precision, with an improved ability to escape from a local optimum.

• Journal of Navigation and Port Research
• /
• v.32 no.3
• /
• pp.223-228
• /
• 2008

As most of small fishing boats made of FRP have been constructed by experience in Korea, some structural safety problems have been occurred occasionally. To improve the structural strength and reduce the costs for construction and operation, optimum design for small fishing boat was carried out in this study. The weight of fishing boat and the main dimensions of structural members are chosen as objective function and design variables, respectively. By the combination of global and local search methods, a hybrid optimization algorithm was developed to escape the local minima and reduce CPU time in analysis procedure, and finite element analysis was performed to determine the constraint parameters at each iteration step in optimization loop. Optimization results were compared with the real existing fishing boat, and the effects of optimum design were examined from points of view; structural strength, material cost, etc.

• Journal of Information Technology Application
• /
• v.1
• /
• pp.173-211
• /
• 1999

The task of classification permeates all walks of life, from business and economics to science and public policy. In this context, nonlinear techniques from artificial intelligence have often proven to be more effective than the methods of classical statistics. The objective of knowledge discovery and data mining is to support decision making through the effective use of information. The automated approach to knowledge discovery is especially useful when dealing with large data sets or complex relationships. For many applications, automated software may find subtle patterns which escape the notice of manual analysis, or whose complexity exceeds the cognitive capabilities of humans. This paper explores the utility of a collaborative learning approach involving integrated models in the preprocessing and postprocessing stages. For instance, a genetic algorithm effects feature-weight optimization in a preprocessing module. Moreover, an inductive tree, artificial neural network (ANN), and k-nearest neighbor (kNN) techniques serve as postprocessing modules. More specifically, the postprocessors act as second0order classifiers which determine the best first-order classifier on a case-by-case basis. In addition to the second-order models, a voting scheme is investigated as a simple, but efficient, postprocessing model. The first-order models consist of statistical and machine learning models such as logistic regression (logit), multivariate discriminant analysis (MDA), ANN, and kNN. The genetic algorithm, inductive decision tree, and voting scheme act as kernel modules for collaborative learning. These ideas are explored against the background of a practical application relating to financial fraud management which exemplifies a binary classification problem.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.101-114
• /
• 2023

In the wartime, aircraft carrying out a mission to strike the enemy deep in the depth are exposed to the risk of being shoot down. As a key combat force in mordern warfare, it takes a lot of time, effot and national budget to train military flight personnel who operate high-tech weapon systems. Therefore, this study studied the path problem of predicting the route of emergency escape from enemy territory to the target point to avoid obstacles, and through this, the possibility of safe recovery of emergency escape military flight personnel was increased. based problem, transforming the problem into a TSP, VRP, and Dijkstra algorithm, and approaching it with an optimization technique. However, if this problem is approached in a network problem, it is difficult to reflect the dynamic factors and uncertainties of the battlefield environment that military flight personnel in distress will face. So, MDP suitable for modeling dynamic environments was applied and studied. In addition, GIS was used to obtain topographic information data, and in the process of designing the reward structure of MDP, topographic information was reflected in more detail so that the model could be more realistic than previous studies. In this study, value iteration algorithms and deterministic methods were used to derive a path that allows the military flight personnel in distress to move to the shortest distance while making the most of the topographical advantages. In addition, it was intended to add the reality of the model by adding actual topographic information and obstacles that the military flight personnel in distress can meet in the process of escape and escape. Through this, it was possible to predict through which route the military flight personnel would escape and escape in the actual situation. The model presented in this study can be applied to various operational situations through redesign of the reward structure. In actual situations, decision support based on scientific techniques that reflect various factors in predicting the escape route of the military flight personnel in distress and conducting combat search and rescue operations will be possible.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.37 no.5
• /
• pp.77-87
• /
• 2000

In this paper, we propose a Over-The-Cell channel routing algorithm for the advanced three-layer process. The proposed algorithm makes the channel routing problem to simplified one and makes use of simulated annealing technique to achieve the global optimal solution. And, a new method to remove the cyclic vertical constraints which are known to be the hardest element in the channel routing problem is proposed, and a way to detect the local minimal solution and escape from it successfully is presented. Futhermore, genetic algorithm based channel router is implemented and comparison is performed with the simulated annealing based one. All algorithms are written in C++ and GUI is made using Motif under Linux environment.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.1
• /
• pp.9-15
• /
• 2009

Ant Colony System is new meta heuristic for hard combinatorial optimization problem. The original ant colony system accomplishes a pheromone updating about only the global optimal path using global updating rule. But, If the global optimal path is not searched until the end condition is satisfied, only pheromone evaporation happens to no matter how a lot of iteration accomplishment. In this paper, the length of the global optimal path does not improved within the limited iterations, we evaluates this state that fall into the local optimum and selects the next node using changed parameters in the state transition rule. This method has effectiveness of the search for a path through diversifications is enhanced by decreasing the value of parameter of the state transition rules for the select of next node, and escape from the local optima is possible. Finally, the performance of Best and Average_Best of proposed algorithm outperforms original ACS.