• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.1
• /
• pp.81-98
• /
• 2013

Since Probabilistic Latent Semantic Analysis (PLSA) and Latent Dirichlet Allocation (LDA) were introduced, many revised or extended topic models have appeared. Due to the intractable likelihood of these models, training any topic model requires to use some approximation algorithm such as variational approximation, Laplace approximation, or Markov chain Monte Carlo (MCMC). Although these approximation algorithms perform well, training a topic model is still computationally expensive given the large amount of data it requires. In this paper, we propose a new method, called non-simultaneous sampling deactivation, for efficient approximation of parameters in a topic model. While each random variable is normally sampled or obtained by a single predefined burn-in period in the traditional approximation algorithms, our new method is based on the observation that the random variable nodes in one topic model have all different periods of convergence. During the iterative approximation process, the proposed method allows each random variable node to be terminated or deactivated when it is converged. Therefore, compared to the traditional approximation ways in which usually every node is deactivated concurrently, the proposed method achieves the inference efficiency in terms of time and memory. We do not propose a new approximation algorithm, but a new process applicable to the existing approximation algorithms. Through experiments, we show the time and memory efficiency of the method, and discuss about the tradeoff between the efficiency of the approximation process and the parameter consistency.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.42 no.4
• /
• pp.194-202
• /
• 2019

Reliability analysis of the components frequently starts with the data that manufacturer provides. If enough failure data are collected from the field operations, the reliability should be recomputed and updated on the basis of the field failure data. However, when the failure time record for a component contains only a few observations, all statistical methodologies are limited. In this case, where the failure records for multiple number of identical components are available, a valid alternative is combining all the data from each component into one data set with enough sample size and utilizing the useful information in the censored data. The ROK Navy has been operating multiple Patrol Killer Guided missiles (PKGs) for several years. The Korea Multi-Function Control Console (KMFCC) is one of key components in PKG combat system. The maintenance record for the KMFCC contains less than ten failure observations and a censored datum. This paper proposes a Bayesian approach with a Dirichlet mixture model to estimate failure time density for KMFCC. Trends test for each component record indicated that null hypothesis, that failure occurrence is renewal process, is not rejected. Since the KMFCCs have been functioning under different operating environment, the failure time distribution may be a composition of a number of unknown distributions, i.e. a mixture distribution, rather than a single distribution. The Dirichlet mixture model was coded as probabilistic programming in Python using PyMC3. Then Markov Chain Monte Carlo (MCMC) sampling technique employed in PyMC3 probabilistically estimated the parameters' posterior distribution through the Dirichlet mixture model. The simulation results revealed that the mixture models provide superior fits to the combined data set over single models.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.1C
• /
• pp.65-71
• /
• 2004

We improved the MLLR speaker adaptation algorithm with reduction of the order of HMM parameters using PCA(Principle Component Analysis) or ICA(Independent Component Analysis). To find a smaller set of variables with less redundancy, we adapt PCA(principal component analysis) and ICA(independent component analysis) that would give as good a representation as possible, minimize the correlations between data elements, and remove the axis with less covariance or higher-order statistical independencies. Ordinary MLLR algorithm needs more than 30 seconds adaptation data to represent higher word recognition rate of SD(Speaker Dependent) models than of SI(Speaker Independent) models, whereas proposed algorithm needs just more than 10 seconds adaptation data. 10 components for ICA and PCA represent similar performance with 36 components for ordinary MLLR framework. So, compared with ordinary MLLR algorithm, the amount of total computation requested in speaker adaptation is reduced by about 1/167 in proposed MLLR algorithm.

• Journal of Internet Computing and Services
• /
• v.9 no.6
• /
• pp.165-177
• /
• 2008

An important issue in Multiagent reinforcement learning is how an agent should learn its optimal policy in a dynamic environment where there exist other agents able to influence its own performance. Most previous works for Multiagent reinforcement learning tend to apply single-agent reinforcement learning techniques without any extensions or require some unrealistic assumptions even though they use explicit models of other agents. In this paper, a Naive Bayesian based policy model of the opponent agent is introduced and then the Multiagent reinforcement learning method using this model is explained. Unlike previous works, the proposed Multiagent reinforcement learning method utilizes the Naive Bayesian based policy model, not the Q function model of the opponent agent. Moreover, this learning method can improve learning efficiency by using a simpler one than other richer but time-consuming policy models such as Finite State Machines(FSM) and Markov chains. In this paper, the Cat and Mouse game is introduced as an adversarial Multiagent environment. And then effectiveness of the proposed Naive Bayesian based policy model is analyzed through experiments using this game as test-bed.

• Asian-Australasian Journal of Animal Sciences
• /
• v.22 no.1
• /
• pp.124-130
• /
• 2009

Estimating genetic interaction effects in animal genomics would be one of the most challenging studies because the phenotypic variation for economically important traits might be largely explained by interaction effects among multiple nucleotide sequence variants under various environmental exposures. Genetic improvement of economic animals would be expected by understanding multi-locus genetic interaction effects associated with economic traits. Most analyses in animal breeding and genetics, however, have excluded the possibility of genetic interaction effects in their analytical models. This review discusses a historical estimation of the genetic interaction and difficulties in analyzing the interaction effects. Furthermore, two recently developed methods for assessing genetic interactions are introduced to animal genomics. One is the restricted partition method, as a nonparametric grouping-based approach, that iteratively utilizes grouping of genotypes with the smallest difference into a new group, and the other is the Bayesian method that draws inferences about the genetic interaction effects based on their marginal posterior distributions and attains the marginalization of the joint posterior distribution through Gibbs sampling as a Markov chain Monte Carlo. Further developing appropriate and efficient methods for assessing genetic interactions would be urgent to achieve accurate understanding of genetic architecture for complex traits of economic animals.

• Journal of Communications and Networks
• /
• v.16 no.1
• /
• pp.81-91
• /
• 2014

This paper proposes hidden-node aware grouping (HAG) algorithm to enhance the performance of institute of electrical and electronics engineers (IEEE) 802.15.4 networks when they undergo either severe collisions or frequent interferences by hidden nodes. According to the degree of measured collisions and interferences, HAG algorithm dynamically transforms IEEE 802.15.4 protocol between a contention algorithm and a contention-limited one. As a way to reduce the degree of contentions, it organizes nodes into some number of groups and assigns each group an exclusive per-group time slot during which only its member nodes compete to grab the channel. To eliminate harmful disruptions by hidden nodes, especially, it identifies hidden nodes by analyzing the received signal powers that each node reports and then places them into distinct groups. For load balancing, finally it flexibly adapts each per-group time according to the periodic average collision rate of each group. This paper also extends a conventional Markov chain model of IEEE 802.15.4 by including the deferment technique and a traffic source to more accurately evaluate the throughput of HAG algorithm under both saturated and unsaturated environments. This mathematical model and corresponding simulations predict with 6%discrepancy that HAG algorithm can improve the performance of the legacy IEEE 802.15.4 protocol, for example, even by 95% in a network that contains two hidden nodes, resulting in creation of three groups.

• Spatial Information Research
• /
• v.2 no.1
• /
• pp.47-51
• /
• 1994

The purpose of the study is to establ ish models of land Cover (use) prediction system for development and management of land resources using remotely sensed data as well as ancillary data in the context of multi-dis¬ciplinary approach in the application to CheJoo Island. The model adopts multi-date processing techniques and is a spatial/temporal land-Cover projection strategy emerged as a synthesis of the probability tra-nsition model and the discrimnant-analys is model. A discriminant modelis applied to all pixels in CheJoo landscape plane to predict the most likely change in land Cover. The probability transition model provides the number of these pixels that will convert to different land Cover in a given future time increment. The syntheric model predicts the future change in land Cover and its volume of pixels in the landscape plane.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.2
• /
• pp.1-9
• /
• 1998

ATM was adopted as the switching and multiplexing technique for BISDN which aims at transmitting traffics with various characteristics in a unified network. To construct these ATM networks, the most important aspect is the design of the switching system with high performance and different service capabilities. In this paepr, we analyze the performance of an input and output queueing switch with preemptive priority which is considered to be most suitable for ATM networks. For the analysis of an input queue, we model each input queue as two separate virtual input queues for each priority class and we approximage them asindependent Geom/Geom/1 queues. And we model a virtual HOL queue which consists of HOL cells of all virtual input queues which have the same output address to obtain the mean service time at each virtual input queue. For the analysis of an output quque, we obtain approximately the arrival process into the output queue from the state of the virtual HOL queue. We use a Markov chain method to analyze these two models and obtain the maximum throughput of the switch and the mean queueing delay of cells. and analysis results are compared with simulation to verify that out model yields accurate results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1579-1602
• /
• 2020

In the design of production system, buffer capacity allocation is a major step. Through polymorphism analysis of production capacity and production capability, this paper investigates a buffer allocation optimization problem aiming at the multi-stage production line including unreliable machines, which is concerned with maximizing the system theoretical production rate and minimizing the system state entropy for a certain amount of buffers simultaneously. Stochastic process analysis is employed to establish Markov models for repairable modular machines. Considering the complex structure, an improved vector UGF (Universal Generating Function) technique and composition operators are introduced to construct the system model. Then the measures to assess the system's multi-state reliability and structural complexity are given. Based on system theoretical production rate and system state entropy, mathematical model for buffer capacity optimization is built and optimized by a specific genetic algorithm. The feasibility and effectiveness of the proposed method is verified by an application of an engine head production line.

• The Korean Journal of Applied Statistics
• /
• v.22 no.1
• /
• pp.115-127
• /
• 2009

The purpose of statistical analyses of array-CGH experiment data is to divide the whole genome into regions of equal copy number, to quantify the copy number in each region and finally to evaluate its significance of being different from two. Several statistical procedures have been proposed which include the circular binary segmentation, and a Gaussian based local regression for detecting break points (GLAD) by estimating a piecewise constant function. We propose in this note a penalized spline regression and its simultaneous confidence band(SCB) approach to evaluate the statistical significance of regions of genetic gain/loss. The region of which the simultaneous confidence band stays above 0 or below 0 can be considered as a region of genetic gain or loss. We compare the performance of the SCB procedure with GLAD and hidden Markov model approaches through a simulation study in which the data were generated from AR(1) and AR(2) models to reflect spatial dependence of the array-CGH data in addition to the independence model. We found that the SCB method is more sensitive in detecting the low level copy number alterations.