• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.5
• /
• pp.684-690
• /
• 2007

We describe our method to predict the direction of conditional probabilistic dependencies between clusters of random variables. Selected variables called 'gateway variables' are used to predict the conditional probabilistic dependency relations between clusters. The direction of conditional probabilistic dependencies between clusters are predicted by finding directed acyclic graph (DAG)-shaped dependency structure between the gateway variables. We show that our method shows meaningful prediction results in determining directions of conditional probabilistic dependencies between clusters.

• Smart Structures and Systems
• /
• v.32 no.1
• /
• pp.61-81
• /
• 2023

This study presents an ensemble learning based Bayesian model updating approach for structural damage diagnosis. In the developed framework, the structure is initially decomposed into a set of substructures. The autoregressive moving average (ARMAX) model is established first for structural damage localization based structural motion equation. The wavelet packet decomposition is utilized to extract the damage-sensitive node energy in different frequency bands for constructing structural surrogate models. Four methods, including Kriging predictor (KRG), radial basis function neural network (RBFNN), support vector regression (SVR), and multivariate adaptive regression splines (MARS), are selected as candidate structural surrogate models. These models are then resampled by bootstrapping and combined to obtain an ensemble model by probabilistic ensemble. Meanwhile, the maximum entropy principal is adopted to search for new design points for sample space updating, yielding a more robust ensemble model. Through the iterations, a framework of surrogate ensemble learning based model updating with high model construction efficiency and accuracy is proposed. The specificities of the method are discussed and investigated in a case study.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.3
• /
• pp.395-400
• /
• 2011

In this paper, we propose an study of the facial emotion recognition considering the dynamical variation of emotional state in facial image sequences. The proposed system consists of two main step: facial image based emotional feature extraction and emotional state classification/recognition. At first, we propose a method for extracting and analyzing the emotional feature region using a combination of Active Shape Model (ASM) and Facial Action Units (FAUs). And then, it is proposed that emotional state classification and recognition method based on Hidden Markov Model (HMM) type of dynamic Bayesian network. Also, we adopt a Harmony Search (HS) algorithm based heuristic optimization procedure in a parameter learning of HMM in order to classify the emotional state more accurately. By using all these methods, we construct the emotion recognition system based on variations of the dynamic facial image sequence and make an attempt at improvement of the recognition performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.10
• /
• pp.3708-3728
• /
• 2021

Indian herbal plants are used in agriculture and in the food, cosmetics, and pharmaceutical industries. Laboratory-based tests are routinely used to identify and classify similar herb species by analyzing their internal cell structures. In this paper, we have applied computer vision techniques to do the same. The original leaf image was preprocessed using the Chan-Vese active contour segmentation algorithm to efface the background from the image by setting the contraction bias as (v) -1 and smoothing factor (µ) as 0.5, and bringing the initial contour close to the image boundary. Thereafter the segmented grayscale image was fed to a leaky capacitance fired neuron model (LCFN), which differentiates between similar herbs by combining different groups of pixels in the leaf image. The LFCN's decay constant (f), decay constant (g) and threshold (h) parameters were empirically assigned as 0.7, 0.6 and h=18 to generate the 1D feature vector. The LCFN time sequence identified the internal leaf structure at different iterations. Our proposed framework was tested against newly collected herbal species of natural images, geometrically variant images in terms of size, orientation and position. The 1D sequence and shape features of aloe, betel, Indian borage, bittergourd, grape, insulin herb, guava, mango, nilavembu, nithiyakalyani, sweet basil and pomegranate were fed into the 5-fold Bayesian regularization neural network (BRNN), K-nearest neighbors (KNN), support vector machine (SVM), and ensemble classifier to obtain the highest classification accuracy of 91.19%.

• Journal of Information Processing Systems
• /
• v.18 no.2
• /
• pp.188-196
• /
• 2022

Recently, neural architecture search (NAS) has received increasing attention as it can replace human experts in designing the architecture of neural networks for different tasks and has achieved remarkable results in many challenging tasks. In this study, a path-based computation neural architecture encoder (PCE) was proposed. Our PCE first encodes the computation of information on each path in a neural network, and then aggregates the encodings on all paths together through an attention mechanism, simulating the process of information computation along paths in a neural network and encoding the computation on the neural network instead of the structure of the graph, which is more consistent with the computational properties of neural networks. We performed an extensive comparison with eight encoding methods on two commonly used NAS search spaces (NAS-Bench-101 and NAS-Bench-201), which included a comparison of the predictive capabilities of performance predictors and search capabilities based on two search strategies (reinforcement learning-based and Bayesian optimization-based) when equipped with different encoders. Experimental evaluation shows that PCE is an efficient encoding method that effectively ranks and predicts neural architecture performance, thereby improving the search efficiency of neural architectures.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.326-338
• /
• 2003

The objective of a neural network design and model selection is to construct an optimal network with a good generalization performance. However, training data include noises, and the number of training data is not sufficient, which results in the difference between the true probability distribution and the empirical one. The difference makes the teaming parameters to over-fit only to training data and to deviate from the true distribution of data, which is called the overfitting phenomenon. The overfilled neural network shows good approximations for the training data, but gives bad predictions to untrained new data. As the complexity of the neural network increases, this overfitting phenomenon also becomes more severe. In this paper, by taking statistical viewpoint, we proposed an integrative process for neural network design and model selection method in order to improve generalization performance. At first, by using the natural gradient learning with adaptive regularization, we try to obtain optimal parameters that are not overfilled to training data with fast convergence. By adopting the natural pruning to the obtained optimal parameters, we generate several candidates of network model with different sizes. Finally, we select an optimal model among candidate models based on the Bayesian Information Criteria. Through the computer simulation on benchmark problems, we confirm the generalization and structure optimization performance of the proposed integrative process of teaming and model selection.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.8
• /
• pp.2948-2963
• /
• 2015

In the field of computer vision, visual surveillance systems have recently become an important research topic. Growth in this area is being driven by both the increase in the availability of inexpensive computing devices and image sensors as well as the general inefficiency of manual surveillance and monitoring. In particular, the ultimate goal for many visual surveillance systems is to provide automatic activity recognition for events at a given site. A higher level of understanding of these activities requires certain lower-level computer vision tasks to be performed. So in this paper, we propose an intelligent activity recognition model that uses a structure learning method and a classification method. The structure learning method is provided as a K2-learning algorithm that generates Bayesian networks of causal relationships between sensors for a given activity. The statistical characteristics of the sensor values and the topological characteristics of the generated graphs are learned for each activity, and then a neural network is designed to classify the current activity according to the features extracted from the multiple sensor values that have been collected. Finally, the proposed method is implemented and tested by using PETS2013 benchmark data.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.10 no.2
• /
• pp.133-144
• /
• 2006

In order to catch out such Bond Strength, the preceding researchers had ever examined the Bond Strength of FRP Plate through their experimentations by setting up of various fluent. However, since the experiment for research on such Bond Strength takes much of expenditure for equipment structure and time-consuming, also difficult to carry out, it is conducting limitedly. This Study purposes to develop the most suitable Artificial Neural Network Model by application of various Neural Network Model and Algorithm to the adhering experiment data of the preceding researchers. Output Layer of Artificial Neural Network Model, and Input Layer of Bond Strength were performed the learning by selection as the variable of the thickness, width, adhered length, the modulus of elasticity, tensile strength, and the compressive strength of concrete, tensile strength, width, respectively. The developed Artificial Neural Network Model has applied Back-Propagation, and its error was learnt to be converged within the range of 0.001. Besides, the process for generalization has dissolved the problem of Over-Fitting in the way of more generalized method by introduction of Bayesian Technique. The verification on the developed Model was executed by comparison with the resulted value of Bond Strength made by the other preceding researchers which was never been utilized to the learning as yet.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1524-1529
• /
• 2005

Particle filter is a kind of conditional density propagation model. Its similar characteristics to both selection and mutation operator of evolutionary strategy (ES) due to its Bayesian inference rule structure, shows better performance than any other tracking algorithms. When a new object is entering the region of interest, particle filter sets which have been swarming around the existing objects have to move and track the new one instantaneously. Moreover, there is another problem that it could not track multiple objects well if they were moving away from each other after having been overlapped. To resolve reinitialization problem, we use competitive-AVQ algorithm of neural network. And we regard interfarme difference (IFD) of background images as potential field and give priority to the particles according to this IFD to track multiple objects independently. In this paper, we showed that the possibility of real-time object tracking as intelligent interfaces by simulating the deformable contour particle filters.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10a
• /
• pp.112-114
• /
• 2004

지능형 로봇 에이전트 기술이 발전하면서 서비스 질을 높이기 위한 방법으로 컨텍스트의 활용성이 부각되고 있다. 하지만 컨텍스트 분류 기술들은 아직까지 초기 개발 단계이며 다양한 방법들이 시도되고 있다. 본 논문에서는 전문가의 지식과 학습된 지식을 함께 적용할 수 있고 사람이 그 내용을 이해하기 유리한 베이지안 네트워크(BN)를 이용한 컨텍스트 분류 방법을 제안한다. 일반적인 BN 구조 학습에 사전 지식 및 방향성, 연결 관계 범위를 부여할 수 있는 제한(Constraint)을 적용한 효율적인 컨텍스트 분류 방법을 소개하고, 몇 가지 비교 실험을 통해 기존 방법에 비해 전문가의 개입이 줄어들고 좀 더 신뢰성 있는 컨텍스트 분류기를 얻을 수 있음을 보인다.