• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.114-114
• /
• 2022

딥러닝의 학습 및 예측성능을 개선하기 위해서는 딥러닝 기법 내 연산과정의 개선과 함께 학습 및 예측에 사용되는 데이터의 전처리 과정이 중요하다. 본 연구에서는 딥러닝의 성능을 개량하기 위해 제안된 메타휴리스틱 최적화 알고리즘-딥러닝 결합모형과 데이터 전처리 기법을 통해 댐의 수위를 예측하였다. 수위예측을 위해 Multi-Layer Perceptron(MLP), 메타휴리스틱 최적화 알고리즘인 Harmony Search(HS)와 딥러닝을 결합한 MLP using a HS(MLPHS) 및 Exponential Bandwidth Harmony Search with Centralized Global Search(EBHS-CGS)와 딥러닝을 결합한MLP using a EBHS-CGS(MLPEBHS)를 통해 댐의 수위를 예측하였다. 메타휴리스틱 최적화 알고리즘-딥러닝 결합모형의 학습 및 예측성능을 개선하기 위해 학습 및 예측을 위한 자료를 기반으로 데이터 전처리기법을 적용하였다. 적용된 데이터 전처리 기법은 정규화, 수위구간별 사상(Event)분리 및 수위 변동에 대한 자료의 구분이다. 수위예측을 위한 대상유역은 금강유역에 위치한 대청댐으로 선정하였다. 대청댐의 수위예측을 위해 대청댐 상류에 위치하는 수위관측소 3개소를 선정하여 수위자료를 취득하였다. 각 수위관측소에서 취득한 수위자료를 입력자료로 설정하였으며, 대청댐의 수위자료를 출력자료로 설정하여 메타휴리스틱 최적화 알고리즘-딥러닝 모형의 학습을 진행하였다. 각 수위관측소 및 대청댐에서 취득한 수위자료는 2010년부터 2020년까지 총 11년의 일 단위 수위자료이며, 2010년부터 2019년까지의 자료를 학습자료로 사용하였으며, 2020년의 자료를 예측 및 검증자료로 사용하였다.

• Smart Structures and Systems
• /
• v.26 no.2
• /
• pp.241-251
• /
• 2020

Due to the undeniable importance of approximating the concrete compressive strength (CSC) in civil engineering, this paper focuses on presenting four novel optimizations of multi-layer perceptron (MLP) neural network, namely artificial bee colony (ABC-MLP), grasshopper optimization algorithm (GOA-MLP), shuffled frog leaping algorithm (SFLA-MLP), and salp swarm algorithm (SSA-MLP) for predicting this crucial parameter. The used dataset consists of 103 rows of information concerning seven influential parameters (cement, slag, water, fly ash, superplasticizer, fine aggregate, and coarse aggregate). In this work, the best-fitted complexity of each ensemble is determined by a population-based sensitivity analysis. The GOA distinguished its self by the least complexity (population size = 50) and emerged as the second time-effective optimizer. Referring to the prediction results, all tested algorithms are able to construct reliable networks. However, the SSA (Correlation = 0.9652 and Error = 1.3939) and GOA (Correlation = 0.9629 and Error = 1.3922) performed more accurately than ABC (Correlation = 0.7060 and Error = 4.0161) and SFLA (Correlation = 0.8890 and Error = 2.5480). Therefore, the SSA-MLP and GOA-MLP can be promising alternatives to laboratorial and traditional CSC evaluative methods.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.10
• /
• pp.831-841
• /
• 1991

A combined model of probabilistic and MLP(multi layer perceptron) model is proposed for the pattern classification of EMG( electromyogram) signals. The MLP model has a problem of not guaranteeing the global minima of error and different quality of approximations to Bayesian probabilities. The probabilistic model is, however, closely related to the estimation error of model parameters and the fidelity of assumptions. A proper combination of these will reduce the effects of the problems and be robust to input variations. Proposed model is able to get the MAP(maximum a posteriori probability) in the probabilistic model by estimating a priori probability distribution using the MLP model adaptively. This method minimize the error probability of the probabilistic model as long as the realization of the MLP model is optimal, and this is a good combination of the probabilistic model and the MLP model for the usage of MLP model reliability. Simulation results show the benefit of the proposed model compared to use the Mlp and the probabilistic model seperately and the average calculation time fro classification is about 50ms in the case of combined motion using an IBM PC 25 MHz 386model.

• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.429-438
• /
• 2006

Neural Networks(NNs) are applied for solving a wide variety of nonlinear problems in several areas, such as image processing, pattern recognition etc. Although NN can be simulated by using software, many potential NN applications required real-time processing. Thus they need to be implemented as hardware. The hardware implementation of multi-layer perceptrons(MLPs) in several kind of NNs usually uses a fixed-point arithmetic due to a simple logic operation and a shorter processing time compared to the floating-point arithmetic. However, the fixed-point arithmetic-based MLP has a drawback which is not able to apply the MLP software that use floating-point arithmetic. We propose a design method for MLPs which has the floating-point arithmetic-based fully-pipelining architecture. It has a processing speed that is proportional to the number of the hidden nodes. The number of input and output nodes of MLPs are generally constrained by given problems, but the number of hidden nodes can be optimized by user experiences. Thus our design method is using optimized number of hidden nodes in order to improve the processing speed, especially in field of a repeated processing such as image processing, pattern recognition, etc.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.210-210
• /
• 2000

In this paper, we have investigated the properties of multi-layer perceptron (MLP) for odour patterns classification and concentration estimation simultaneously. When the MLP may be has a fast convergence speed with small error and excellent mapping ability for classification, it can be possible to use for classification and concentration prediction of volatile chemicals simultaneously. However, the conventional MLP, which is back-Propagation of error based on the steepest descent method, was difficult to use for odour classification and concentration estimation simultaneously, because it is slow to converge and may fall into the local minimum. We adapted the Levenberg-Marquardt(LM) algorithm [4,5] having advantages both the steepest descent method and Gauss-Newton method instead of the conventional steepest descent method for the simultaneous classification and concentration estimation of odours. And, We designed the artificial odour sensing system(Electronic Nose) and applied LM-BP algorithm for classification and concentration prediction of VOC gases.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.261-262
• /
• 2006

According to the previous work, various well known methods including maximum a posteriori probability classifier (MAP) and multi layer perceptron networks classifier (MLP) showed competitive results. Since even the small errors often leads to a fatal result, we investigated the method that reduces classification error perfectly by screening out some ambiguous patterns. Those ambiguous patterns can be examined by routine biopsy. We incorporated an ambiguous category in MAP and MLP. Classification results involving 216 spectra gave 100% sensitivity for the case of MLP.

• Annual Conference of KIPS
• /
• 2018.10a
• /
• pp.920-922
• /
• 2018

This paper describes a method to classify simple circular artificial markers on surfaces of a box on the back of hand to detect the pose of user's hand for VR/AR applications by using a Leap Motion camera and two IMU sensors. One IMU sensor is located in the box and the other IMU sensor is fixed with the camera. Multi-layer Perceptron (MLP) algorithm is adopted to classify artificial markers on each surface tracked by the camera using IMU sensor data. It is experimented successfully in real-time, 70Hz, under PC environments.

• Steel and Composite Structures
• /
• v.45 no.6
• /
• pp.877-894
• /
• 2022

Geopolymer concrete (GPC) has emerged as a feasible choice for construction materials as a result of the environmental issues associated with the production of cement. The findings of this study contribute to the development of machine learning methods for estimating the properties of eco-friendly concrete to help reduce CO₂ emissions in the construction industry. The compressive strength (f_c) of GPC is predicted using artificial intelligence approaches in the present study when ground granulated blast-furnace slag (GGBS) is substituted with natural zeolite (NZ), silica fume (SF), and varying NaOH concentrations. For this purpose, two machine learning methods multi-layer perceptron (MLP) and radial basis function (RBF) were considered and hybridized with arithmetic optimization algorithm (AOA), and grey wolf optimization algorithm (GWO). According to the results, all methods performed very well in predicting the f_c of GPC. The proposed AOA - MLP might be identified as the outperformed framework, although other methodologies (AOA - RBF, GWO - RBF, and GWO - MLP) were also reliable in the fc of GPC forecasting process.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.1116-1119
• /
• 2021

2018 년 Google 사의 사전 학습된 언어 인공지능 BERT 를 기점으로, 자연어 처리 학계는 주요 구조를 유지한 채 경쟁적으로 모델을 대형화하는 방향으로 발전했다. 그 결과, 오늘날 자연어 인공지능은 거대 사기업과 그에 준하는 컴퓨팅 자원을 소유한 연구 단체만의 전유물이 되었다. 본 논문에서는 다층 퍼셉트론을 병렬적으로 배열해 자연어 인공지능을 제작하는 기법의 모델을 제안하고, 이를 적용한'조건부 게이트 다층 퍼셉트론 모델(SG-MLP)'을 구현하고 그 결과를 비교 관찰하였다. SG-MLP 는 BERT 의 20%에 해당하는 사전 학습량만으로 다수의 지표에서 그것과 준하는 성능을 보였고, 동일한 과제에 대해 더 적은 연산 비용을 소요한다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.22 no.10
• /
• pp.981-989
• /
• 2011

In this paper, we performed radar target classification for the five scale models using ultra-wideband measured signal. In order to compare the performance, the 2 GHz(2~4 GHz), 4 GHz(2~6 GHz), and 6 GHz(2~8 GHz) bandwidth were used. Short time Fourier transform(STFT) and continuous wavelet transform(CWT) are used for target feature extraction. Extracted feature vectors are used as input for the multi-layerd perceptron(MLP) neural network classifier. The results show that as the bandwidth is wider, the performance is better.