• Proceedings of the Optical Society of Korea Conference
• /
• 1991.06a
• /
• pp.35-40
• /
• 1991

In this paper, a new adaptive learning algorithm, Training by Adaptive Gain (TAG) for optical implementation of large-sized neural networks has been developed and its electro-optical implementation for 2-dimensional input and output neurons has been demostrated. The 4-dimensional global fixed interconnections and 2-dimensional adaptive gain-controls are implemented by multi-facet computer generated holograms and LCTV spatial light modulators, respectively. When the input signals pass through optical system to the output classifying layer, the TAG adaptive learning algorithm is implemented by a personal computer. The system classifies three 5$\times$5 input patterns correctly.

• The Journal of the Korea Contents Association
• /
• v.22 no.7
• /
• pp.55-62
• /
• 2022

The perovskite solar cell is an active part of research in renewable energy fields such as solar energy, wind, hydroelectric power, marine energy, bioenergy, and hydrogen energy to replace fossil fuels such as oil, coal, and natural gas, which will gradually disappear as power demand increases due to the increase in use of the Internet of Things and Virtual environments due to the 4th industrial revolution. The perovskite solar cell is a solar cell device using an organic-inorganic hybrid material having a perovskite structure, and has advantages of replacing existing silicon solar cells with high efficiency, low cost solutions, and low temperature processes. In order to optimize the light absorption layer thin film predicted by the existing empirical method, reliability must be verified through device characteristics evaluation. However, since it costs a lot to evaluate the characteristics of the light-absorbing layer thin film device, the number of tests is limited. In order to solve this problem, the development and applicability of a clear and valid model using machine learning or artificial intelligence model as an auxiliary means for optimizing the light absorption layer thin film are considered infinite. In this study, to estimate the light absorption layer thin-film optimization of perovskite solar cells, the regression models of the support vector machine's linear kernel, R.B.F kernel, polynomial kernel, and sigmoid kernel were compared to verify the accuracy difference for each kernel function.

• Journal of the Korea Convergence Society
• /
• v.9 no.5
• /
• pp.1-6
• /
• 2018

In smart grid an accurate load forecasting is crucial in planning resources, which aids in improving its operation efficiency and reducing the dynamic uncertainties of energy systems. Research in this area has included the use of shallow neural networks and other machine learning techniques to solve this problem. Recent researches in the field of computer vision and speech recognition, have shown great promise for Deep Neural Networks (DNN). To improve the performance of daily electric peak load forecasting the paper presents a new deep neural network model which has the architecture of two multi-layer neural networks being serially connected. The proposed network model is progressively pre-learned layer by layer ahead of learning the whole network. For both one day and two day ahead peak load forecasting the proposed models are trained and tested using four years of hourly load data obtained from the Korea Power Exchange (KPX).

• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.758-765
• /
• 2021

Recently, various research on the application of deep learning to video encoding for enhancing coding efficiency are being actively studied. This paper proposes a deep learning based intra prediction which uses multiple models by extending Matrix-based Intra Prediction(MIP) that is a neural network-based technology adopted in VVC. It also presents an efficient learning method for the multi-model intra prediction. To evaluate the performance of the proposed method, we integrated the VVC MIP and the proposed fully connected layer based multi-model intra prediction into HEVC reference software, HM16.19 as an additional intra prediction mode. As a result of the experiments, the proposed method can obtain bit-saving coding gain up to 0.47% and 0.19% BD-rate, respectively, compared to HM16.19 and VVC MIP.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.5
• /
• pp.598-603
• /
• 2007

To solve the rule explosion problem in multi-input fuzzy system, a method of converting a given fuzzy system to 2 layered hierarchical fuzzy system has been reported, where at the 1st layer, linearly independent fuzzy rule vectors generated from the given fuzzy system are used and, at the 2nd layer, linear combinations of these independent fuzzy rule vectors are used. In this paper, the steapest descent algorithm is presented to learn the fuzzy rule vectors and related coefficients for the equivalent 2 layered hierarchical structure. By simulation of learning of ball and beam control system, the feasibility of proposed learning scheme is shown.

• Structural Engineering and Mechanics
• /
• v.84 no.3
• /
• pp.323-335
• /
• 2022

The main idea of the framework is to seamlessly combine a reasonably accurate and fast surrogate model with the importance sampling strategy. Developing a surrogate model for predicting structures' dynamic responses is challenging because it involves high-dimensional inputs and outputs. For this purpose, a novel surrogate model based on cutting-edge deep learning architectures specialized for capturing temporal relationships within time-series data, namely Long-Short term memory layer and Transformer layer, is designed. After being properly trained, the surrogate model could be utilized in place of the finite element method to evaluate structures' responses without requiring any specialized software. On the other hand, the importance sampling is adopted to reduce the number of calculations required when computing the failure probability by drawing more relevant samples near critical areas. Thanks to the portability of the trained surrogate model, one can integrate the latter with the Importance sampling in a straightforward fashion, forming an efficient framework called TTIS, which represents double advantages: less number of calculations is needed, and the computational time of each calculation is significantly reduced. The proposed approach's applicability and efficiency are demonstrated through three examples with increasing complexity, involving a 1D beam, a 2D frame, and a 3D building structure. The results show that compared to the conventional Monte Carlo simulation, the proposed method can provide highly similar reliability results with a reduction of up to four orders of magnitudes in time complexity.

• Journal of Information Processing Systems
• /
• v.19 no.6
• /
• pp.842-857
• /
• 2023

Currently, most sentiment classification models on microblogging platforms analyze sentence parts of speech and emoticons without comprehending users' emotional inclinations and grasping moral nuances. This study proposes a hybrid sentiment analysis model. Given the distinct nature of microblog comments, the model employs a combined stop-word list and word2vec for word vectorization. To mitigate local information loss, the TextCNN model, devoid of pooling layers, is employed for local feature extraction, while BiLSTM is utilized for contextual feature extraction in deep learning. Subsequently, microblog comment sentiments are categorized using a classification layer. Given the binary classification task at the output layer and the numerous hidden layers within BiLSTM, the Tanh activation function is adopted in this model. Experimental findings demonstrate that the enhanced TextCNN-BiLSTM model attains a precision of 94.75%. This represents a 1.21%, 1.25%, and 1.25% enhancement in precision, recall, and F1 values, respectively, in comparison to the individual deep learning models TextCNN. Furthermore, it outperforms BiLSTM by 0.78%, 0.9%, and 0.9% in precision, recall, and F1 values.

• Journal of Korea Multimedia Society
• /
• v.24 no.10
• /
• pp.1414-1424
• /
• 2021

In this paper, the importance of input factors of a DNN (Deep Neural Network) PM_2.5 forecasting model using LRP(Layer-wise Relevance Propagation) is analyzed, and forecasting performance is improved. Input factor importance analysis is performed by dividing the learning data into time and PM_2.5 concentration. As a result, in the low concentration patterns, the importance of weather factors such as temperature, atmospheric pressure, and solar radiation is high, and in the high concentration patterns, the importance of air quality factors such as PM_2.5, CO, and NO₂ is high. As a result of analysis by time, the importance of the measurement factors is high in the case of the forecast for the day, and the importance of the forecast factors increases in the forecast for tomorrow and the day after tomorrow. In addition, date, temperature, humidity, and atmospheric pressure all show high importance regardless of time and concentration. Based on the importance of these factors, the LRP_DNN prediction model is developed. As a result, the ACC(accuracy) and POD(probability of detection) are improved by up to 5%, and the FAR(false alarm rate) is improved by up to 9% compared to the previous DNN model.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.847-851
• /
• 2004

In the original mixture density network(MDN), which was introduced by Bishop and Nabney, the parameters of the conditional probability density function are represented by the output vector of a single multi-layer perceptron. Among the recent modification of the MDNs, there is the so-called modified mixture density network, in which each of the priors, conditional means, and covariances is represented via an independent multi-layer perceptron. In this paper, we consider a further simplification of the modified MDN, in which the conditional means are linear with respect to the input variable together with the development of the MATLAB program for the simplification. In this paper, we first briefly review the original mixture density network, then we also review the modified mixture density network in which independent multi-layer perceptrons play an important role in the learning for the parameters of the conditional probability, and finally present a further modification so that the conditional means are linear in the input. The applicability of the presented method is shown via an illustrative simulation example.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.7
• /
• pp.1706-1725
• /
• 2024

The unique U-shaped structure of U-Net network makes it achieve good performance in image segmentation. This network is a lightweight network with a small number of parameters for small image segmentation datasets. However, when the medical image to be segmented contains a lot of detailed information, the segmentation results cannot fully meet the actual requirements. In order to achieve higher accuracy of medical image segmentation, a novel improved U-Net network architecture called multi-scale encoder-decoder U-Net+ (MEDU-Net+) is proposed in this paper. We design the GoogLeNet for achieving more information at the encoder of the proposed MEDU-Net+, and present the multi-scale feature extraction for fusing semantic information of different scales in the encoder and decoder. Meanwhile, we also introduce the layer-by-layer skip connection to connect the information of each layer, so that there is no need to encode the last layer and return the information. The proposed MEDU-Net+ divides the unknown depth network into each part of deconvolution layer to replace the direct connection of the encoder and decoder in U-Net. In addition, a new combined loss function is proposed to extract more edge information by combining the advantages of the generalized dice and the focal loss functions. Finally, we validate our proposed MEDU-Net+ MEDU-Net+ and other classic medical image segmentation networks on three medical image datasets. The experimental results show that our proposed MEDU-Net+ has prominent superior performance compared with other medical image segmentation networks.