• Journal of the Korean Statistical Society
• /
• v.29 no.4
• /
• pp.407-422
• /
• 2000

This article is concerned with the selecting predictor variables to be included in building a class of binary response t-link regression models where both probit and logistic regression models can e approximately taken as members of the class. It is based on a modification of the stochastic search variable selection method(SSVS), intended to propose and develop a Bayesian procedure that used probabilistic considerations for selecting promising subsets of predictor variables. The procedure reformulates the binary response t-link regression setup in a hierarchical truncated normal mixture model by introducing a set of hyperparameters that will be used to identify subset choices. In this setup, the most promising subset of predictors can be identified as that with highest posterior probability in the marginal posterior distribution of the hyperparameters. To highlight the merit of the procedure, an illustrative numerical example is given.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.1
• /
• pp.87-98
• /
• 2002

A competitive coevolutionary algorithm is a probabilistic search method that imitates coevolution process through evolutionary arms race. The algorithm has been used to solve adversarial problems. In the algorithms, the selection of competitors is needed to evaluate the fitness of an individual. The goal of this study is to compare and analyze several competition strategies in terms of solution quality, convergence speed, balance between competitive coevolving species, population diversity, etc. With two types of test-bed problems, game problems and solution-test problems, extensive experiments are carried out. In the game problems, sampling strategies based on fitness have a risk of providing bad solutions due to evolutionary unbalance between species. On the other hand, in the solution-test problems, evolutionary unbalance does not appear in any strategies and the strategies using information about competition results are efficient in solution quality. The experimental results indicate that the tournament competition can progress an evolutionary arms race and then is successful from the viewpoint of evolutionary computation.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.4E
• /
• pp.21-27
• /
• 1996

In this paper, we propose a discriminative training algorithm for the stochastic segment model (SSM) in continuous speech recognition. As the SSM is usually trained by maximum likelihood estimation (MLE), a discriminative training algorithm is required to improve the recognition performance. Since the SSM does not assume the conditional independence of observation sequence as is done in hidden Markov models (HMMs), the search space for decoding an unknown input utterance is increased considerably. To reduce the computational complexity and starch space amount in an iterative training algorithm for discriminative SSMs, a hybrid architecture of SSMs and HMMs is programming using HMMs. Given the segment boundaries, the parameters of the SSM are discriminatively trained by the minimum error classification criterion based on a generalized probabilistic descent (GPD) method. With the discriminative training of the SSM, the word error rate is reduced by 17% compared with the MLE-trained SSM in speaker-independent continuous speech recognition.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.264-266
• /
• 2009

A skill is defined as the special ability to do something well, especially as acquired by learning and practice. To learn a skill, a Bayesian network model for representing the skill is first learned. We will regard the Bayesian network for a skill as an affordance. We propose a soft Behavior Motivation(BM) switch as a method for ordering affordances to accomplish a task. Then, a skill is constructed as a combination of an affordance and a soft BM switch. To demonstrate the validity of our proposed method, some experiments were performed with GENIBO(Pet robot) performing a task using skills of Search-a-target-object, Approach-a-target-object, Push-up-in front of -a-target-object.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.420-426
• /
• 1998

Evolutionary algorithms are probabilistic optimization algorithms based on the model of natural evolution. Recently the efforts to improve the performance of evolutionary algorithms have been made extensively. In this paper, we introduce the research for improving the convergence rate and search faculty of evolution algorithms by using reinforcement learning. After providing an introduction to evolution algorithms and reinforcement learning, we present adaptive genetic algorithms, reinforcement genetic programming, and reinforcement evolution strategies which are combined with reinforcement learning. Adaptive genetic algorithms generate mutation probabilities of each locus by interacting with the environment according to reinforcement learning. Reinforcement genetic programming executes crossover and mutation operations based on reinforcement and inhibition mechanism of reinforcement learning. Reinforcement evolution strategies use the variances of fitness occurred by mutation to make the reinforcement signals which estimate and control the step length.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.52 no.1
• /
• pp.36-43
• /
• 2003

This paper presents an optimal tuning method for Fuzzy Logic Controller (FLC) of current controller for HVDC using Genetic Algorithm(GA). GA is probabilistic search method based on genetics and evolution theory. The scaling factors of FLC are tuned by using real-time GA. The proposed tuning method is applied to the scaled-down HVDC simulator at Korea Electrotechnology Research Institute(KERI). Experimental result shows that disturbances are well-damped and the dynamic performances of FLC have the better responses than those of PI controller for small and large disturbances such as ULTC tap change, reference DC current change and DC ground fault.

• Journal of Korean Institute of Industrial Engineers
• /
• v.24 no.4
• /
• pp.559-569
• /
• 1998

Cellular manufacturing system (CMS) has been recognized as an alternative to improve manufacturing productivity in conventional batch-type manufacturing systems through reducing set-up times, work-in-process inventories and throughput times by means of group technology. Most of the studies on the design of CMS assumed that each part has a unique process plan, and that its demand is known as a deterministic value despite of the probabilistic nature of the real world problems. This study suggests an approach for designing CMS, considering both alternative process plans and uncertain demand. A mathematical model is presented to show how to minimize the expected amortized and operating costs satisfying these two relaxations. Four heuristic algorithms are developed based on tabu search which is well suited for getting an optimal or near-optimal solution. Example problems are carried out to illustrate the heuristic algorithms and each of them is compared with the deterministic counterpart.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.2
• /
• pp.166-172
• /
• 2014

Image classification is an important problem in computer vision. However, it is a very challenging problem due to the variability, ambiguity and scale change that exists in images. In this paper, we propose a method of a hypergraph based modeling can consider the higher-order relationships of semantic attributes of a scene image and apply it to a scene image classification. In order to generate the hypergraph optimized for specific scene category, we propose a novel search method based on a probabilistic subspace method and also propose a method to aggregate the expression values of the member semantic attributes that belongs to the searched subsets based on a linear transformation method via likelihood based estimation. To verify the superiority of the proposed method, we showed that the discrimination power of the feature vector generated by the proposed method is better than existing methods through experiments. And also, in a scene classification experiment, the proposed method shows a competitive classification performance compared with the conventional methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.6
• /
• pp.1294-1301
• /
• 2014

In this paper, separate probabilistic distribution models for voiced and unvoiced speech are estimated and utilized to improve speaker recognition performance. Also, in addition to the conventional mel-frequency cepstral coefficient, skewness, kurtosis, and harmonic-to-noise ratio are extracted and used for voiced speech intervals. Two kinds of scores for voiced and unvoiced speech are linearly fused with the optimal weight found by exhaustive search. The performance of the proposed speaker recognizer is compared with that of the conventional recognizer which uses mel-frequency cepstral coefficient and a unified probabilistic distribution function based on the Gassian mixture model. Experimental results show that the lower the number of Gaussian mixture, the greater the performance improvement by the proposed algorithm.

• Journal of IKEEE
• /
• v.1 no.1 s.1
• /
• pp.31-46
• /
• 1997

In this paper, a parallel Optimal Best-First search Branch-and-Bound(B&B) algorithm(pobs) is designed and evaluated for MIN-based multiprocessor systems. The proposed algorithm decomposes a problem into G subproblems, where each subproblem is processed on a group of P processors. Each processor group uses tile sub-Global Best-First search technique to find a local solution. The local solutions are broadcasted through the network to compute the global solution. This broadcast provides not only the comparison of G local solutions but also the load balancing among the processor groups. A performance analysis is then conducted to estimate the speed-up of the proposed parallel B&B algorithm. The analytical model is developed based on the probabilistic properties of the B&B algorithm. It considers both the computation time and communication overheads to evaluate the realistic performance of the algorithm under the parallel processing environment. In order to validate the proposed evaluation model, the simulation of the parallel B&B algorithm on a MIN-based system is carried out at the same time. The results from both analysis and simulation match closely. It is also shown that the proposed Optimal Best-First search B&B algorithm performs better than other reported schemes with its various advantageous features such as: less subproblem evaluations, prefer load balancing, and limited scope of remote communication.