• The KIPS Transactions:PartB
• /
• v.14B no.7
• /
• pp.541-548
• /
• 2007

Due to the mathematical convergence property, Simulated Annealing (SA) has been one of the most popular optimization algorithms. However, because of its problem of slow convergence in the practical use, many variations of SA like Fast SA (FSA) have been developed for faster convergence. In this paper, we propose and prove that Greedy SA (GSA) also finds the global optimum in probability in the continuous space optimization problems. Because the greedy selection does not allow the cost to become worse, GSA is expected to have faster convergence than the conventional FSA that uses Metropolis selection. In the computer simulation, the proposed method is shown to have as good performance as FSA with Metropolis selection in the viewpoints of the convergence speed and the quality of the found solution. Furthermore, the greedy selection does not concern the cost value itself but uses only dominance of the costs of solutions, which makes GSA invariant to the problem scaling.

• The Korean Journal of Applied Statistics
• /
• v.14 no.2
• /
• pp.321-332
• /
• 2001

본 논문에서는 반응변수가 두 가지의 값을 갖는 회귀분석에 적용할 수 있는 사영추적회귀를 고려하였다. 회귀모형에 필요한 설명변수들의 선형결합이 하나이고 연결함수의 형태를 사전에 알지 못한다는 가정하에서 모의담금질 기법을 이용하여 모형에 필요한 선형결합을 찾는 알고리즘을 제시하였다. 이진 반응변수의 경우에는 평활모수의 값에 따라 잔차이탈도함수의 반응표면이 단봉의 형태를 갖지 않는 경우가 있어 비동질적 마코프체인을 이용한 모의담금질 기법을 적용하면 효율적으로 선형결합을 탐색할 수 있다.

• The KIPS Transactions:PartB
• /
• v.14B no.5
• /
• pp.377-382
• /
• 2007

This paper proposes a training algorithm for global optimization of the parameters of radial basis function networks. Since conventional training algorithms usually perform only local optimization, the performance of the network is limited and the final network significantly depends on the initial network parameters. The proposed hybrid simulated annealing algorithm performs global optimization of the network parameters by combining global search capability of simulated annealing and local optimization capability of gradient-based algorithms. Via experiments for function approximation problems, we demonstrate that the proposed algorithm can find networks showing better training and test performance and reduce effects of the initial network parameters on the final results.

• Tunnel and Underground Space
• /
• v.15 no.4 s.57
• /
• pp.275-287
• /
• 2005

At the planning and design stages of a development of underground space or tunneling project, the information regarding ground conditions is very important to enhance economical efficiency and overall safety In general, the information can be expressed using RMR or Q-system and with the geophysical exploration image. RMR or Q-system can provide direct information of rock mass in a local scale for the design scheme. Oppositely, the image of geophysical exploration can provide an exthaustive but indirect information. These two types of the information have inherent uncertainties from various sources and are given in different scales and with their own physical meanings. Recently, RMR has been estimated in unsampled areas based on given data using geostatistical methods like Kriging and conditional simulation. In this study, simulated annealing(SA) is applied to overcome the shortcomings of Kriging methods or conditional simulations just using a primary variable. Using this technique, RMR and the image of geophysical exploration can be integrated to construct the spatial distribution of RM and to evaluate its uncertainty. The SA method was applied to solve an optimization problem with constraints. We have suggested the practical procedure of the SA technique for the uncertainty evaluation of RMR and also demonstrated this technique through an application, where it was used to identify the spatial distribution of RMR and quantify the uncertainty. For a geotechnical application, the objective functions of SA are defined using statistical models of RMR and the correlations between RMR and the reference image. The applicability and validity of this application are examined and then the result of uncertainty evaluation can be used to optimize the tunnel layout.

• The KIPS Transactions:PartB
• /
• v.14B no.7
• /
• pp.523-530
• /
• 2007

This paper proposes a new stochastic optimization algorithm for hidden Markov models (HMMs) used as a recognizer of automatic lipreading. The proposed method combines a global stochastic optimization method, the simulated annealing technique, and the local optimization method, which produces fast convergence and good solution quality. We mathematically show that the proposed algorithm converges to the global optimum. Experimental results show that training HMMs by the method yields better lipreading performance compared to the conventional training methods based on local optimization.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.53-60
• /
• 2008

This paper proposes a new multiobjective optimization method for discriminative training of hidden Markov models (HMMs) used as the recognizer for automatic lipreading. While the conventional Baum-Welch algorithm for training HMMs aims at maximizing the probability of the data of a class from the corresponding HMM, we define a new training criterion composed of two minimization objectives and develop a global optimization method of the criterion based on simulated annealing. The result of a speaker-dependent recognition experiment shows that the proposed method improves performance by the relative error reduction rate of about 8% in comparison to the Baum-Welch algorithm.

• Journal of the Korea Society for Simulation
• /
• v.32 no.1
• /
• pp.23-34
• /
• 2023

The purpose of this paper is to review global stochastic optimization algorithms(GSOA) in case binary response experimentation is used and to compare the performances of them. GSOAs utilise estimator of probability of success $\^p$ instead of population probability of success p, since p is unknown and only known by its estimator which has stochastic characteristics. Hill climbing algorithm algorithm, simple random search, random search with random restart, random optimization, simulated annealing and particle swarm algorithm as a population based algorithm are considered as global stochastic optimization algorithms. For the purpose of comparing the algorithms, two types of test functions(one is simple uni-modal the other is complex multi-modal) are proposed and Monte Carlo simulation study is done to measure the performances of the algorithms. All algorithms show similar performances for simple test function. Less greedy algorithms such as Random optimization with Random Restart and Simulated Annealing, Particle Swarm Optimization(PSO) based on population show much better performances for complex multi-modal function.

• 베이커리
• /
• no.7 s.456
• /
• pp.124-126
• /
• 2006

오는 7월7일부터 8일까지 이틀 동안 미국 애리조나주 피닉스에서 세계 제과인이 한 자리에 모여 열띤 경합을 펼치게 된다. 2006 월드페이스트리팀챔피언십에 출전하는 세 사람의 한국대표선수는 지난 7개월 동안의 쉼없는 담금질을 거치면서 붉은 악마처럼 악착같은 패기를, 월드컵 태극전사처럼 노련한 기술을 지니게 됐다.

• Communications for Statistical Applications and Methods
• /
• v.15 no.4
• /
• pp.633-642
• /
• 2008

Variable selection algorithm for partial least square regression using penalty function is proposed. We use the fact that usual partial least square regression problem can be expressed as a maximization problem with appropriate constraints and we will add penalty function to this maximization problem. Then simulated annealing algorithm can be used in searching for optimal solutions of above maximization problem with penalty functions added. The HARD penalty function would be suggested as the best in several aspects. Illustrations with real and simulated examples are provided.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.651-652
• /
• 2008

In this paper, we suggest a new multiobjective optimization algorithm which is based on the simulated annealing(SA) method. The proposed algorithm uses population-based simulated annealing and adapts elitism in the process of selection.