• ETRI Journal
• /
• v.17 no.1
• /
• pp.11-22
• /
• 1995

This paper proposes a modified error function to improve the error back-propagation (EBP) algorithm for multi-Layer perceptrons (MLPs) which suffers from slow learning speed. It can also suppress over-specialization for training patterns that occurs in an algorithm based on a cross-entropy cost function which markedly reduces learning time. In the similar way as the cross-entropy function, our new function accelerates the learning speed of the EBP algorithm by allowing the output node of the MLP to generate a strong error signal when the output node is far from the desired value. Moreover, it prevents the overspecialization of learning for training patterns by letting the output node, whose value is close to the desired value, generate a weak error signal. In a simulation study to classify handwritten digits in the CEDAR [1] database, the proposed method attained 100% correct classification for the training patterns after only 50 sweeps of learning, while the original EBP attained only 98.8% after 500 sweeps. Also, our method shows mean-squared error of 0.627 for the test patterns, which is superior to the error 0.667 in the cross-entropy method. These results demonstrate that our new method excels others in learning speed as well as in generalization.

• Journal of Astronomy and Space Sciences
• /
• v.34 no.2
• /
• pp.127-138
• /
• 2017

The current study designs the mission orbit of the lunar CubeSat spacecraft to measure the lunar local magnetic anomaly. To perform this mission, the CubeSat will impact the lunar surface over the Reiner Gamma swirl on the Moon. Orbit analyses are conducted comprising ${\Delta}V$ and error propagation analysis for the CubeSat mission orbit. First, three possible orbit scenarios are presented in terms of the CubeSat's impacting trajectories. For each scenario, it is important to achieve mission objectives with a minimum ${\Delta}V$ since the CubeSat is limited in size and cost. Therefore, the ${\Delta}V$ needed for the CubeSat to maneuver from the initial orbit toward the impacting trajectory is analyzed for each orbit scenario. In addition, error propagation analysis is performed for each scenario to evaluate how initial errors, such as position error, velocity error, and maneuver error, that occur when the CubeSat is separated from the lunar orbiter, eventually affect the final impact position. As a result, the current study adopts a CubeSat release from the circular orbit at 100 km altitude and an impact slope of $15^{\circ}$, among the possible impacting scenarios. For this scenario, the required ${\Delta}V$ is calculated as the result of the ${\Delta}V$ analysis. It can be used to practically make an estimate of this specific mission's fuel budget. In addition, the current study suggests error constraints for ${\Delta}V$ for the mission.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.113-117
• /
• 1998

We propose a way to show the inherent learning complexity for the multilayer perceptron. We display the solution space and the error surfaces on the input space of a single neuron with two inputs. The evolution of its weights will follow one of the two error surfaces. We observe that when we use the back-propagation(BP) learning algorithm (1), the wight cam not jump to the lower error surface due to the implicit continuity constraint on the changes of weight. The self-relaxation approach is to explicity find out the best combination of all neurons' two error surfaces. The time complexity of training a multilayer perceptron by self-relaxationis exponential to the number of neurons.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.853-858
• /
• 1993

The error back-propagation(BP) algorithm is widely used for finding optimum weights of multi-layer neural networks. However, the critical drawback of the BP algorithm is its slow convergence of error. The major reason for this slow convergence is the premature saturation which is a phenomenon that the error of a neural network stays almost constant for some period time during learning. An inappropriate selections of initial weights cause each neuron to be trapped in the premature saturation state, which brings in slow convergence speed of the multi-layer neural network. In this paper, to overcome the above problem, Micro-Genetic algorithms(.mu.-GAs) which can allow to find the near-optimal values, are used to select the proper weights and slopes of activation function of neurons. The effectiveness of the proposed algorithms will be demonstrated by some computer simulations of two d.o.f planar robot manipulator.

• Journal of Magnetics
• /
• v.22 no.1
• /
• pp.34-42
• /
• 2017

Material defects in the Steam Generator Tubes (SGT) of sodium cooled fast breeder reactor (PFBR) can lead to leakage of water into sodium. The water and sodium reaction will lead to major accidents. Therefore, the examination of steam generator tubes for the early detection of defects is an important requirement for safety and economic considerations. In this work, the Magnetic Flux Leakage (MFL) based Non Destructive Testing (NDT) technique is used to perform the defect detection process. The rectangular notch defects on the outer surface of steam generator tubes are modeled using COMSOL multiphysics 4.3a software. The obtained MFL images are de-noised to improve the integrity of flaw related information. Grey Level Co-occurrence Matrix (GLCM) features are extracted from MFL images and taken as input parameter to train the neural network. A comparative study on characterization have been carried out using feed-forward back propagation (FFBP) and cascade-forward back propagation (CFBP) algorithms. The results of both algorithms are evaluated with Mean Square Error (MSE) as a prediction performance measure. The average percentage error for length, depth and width are also computed. The result shows that the feed-forward back propagation network model performs better in characterizing the defects.

• ETRI Journal
• /
• v.29 no.5
• /
• pp.559-568
• /
• 2007

This paper presents a new distortion control scheme with a simple estimation model for the propagation errors incurred by dropping some parts of the bitstream in a frame dropping-coefficient dropping (FD-CD) transcoder. The primary goal of this paper is to facilitate bit-rate conversions and rate-distortion controls in the compressed domain without introducing a full decoding and reencoding system in the pixel domain. First, the error propagation behavior over several frame sequences due to coefficient dropping is investigated on the basis of statistical and empirical properties. Then, such properties are used to develop a simple estimation model for the CD distortion accounting for the characteristics of the underlying coded-frame. Finally, the proposed estimation model allows us to determine the amount of coefficient dropping and to effectively allocate rate-distortions into coded-frames. Experimental results show that the proposed estimation model accurately describes the characteristics of propagation errors adaptively in the compressed domain and can be easily applied to distortion control over different kinds of video sequences.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.6
• /
• pp.67-75
• /
• 1998

Although the nCE(n-th order cross-entropy) error function resolves the incorrect saturation problem of conventional EBP(error back-propagation) algorithm, the performance of MLP's (multilayer perceptrons) trained using the nCE function depends heavily on the order of the nCE function. In this paper, we propose an adaptive learning rate to make the performance of MLP's insensitive to the order of the nCE error. Additionally, we propose a limited error signal of output node to prevent unstable learning due to the adaptive learning rate. The effectiveness of the proposed method is demonstrated in simulations of handwritten digit recognition and thyroid diagnosis tasks.

• The KIPS Transactions:PartD
• /
• v.14D no.7
• /
• pp.809-818
• /
• 2007

In this thesis, the color information is used to extract the traffic sign territory, and for recognizing the extracted image, it proposes the traffic sign recognition system that applies the error back propagation algorithm. The proposed method analyzes the color of traffic sign to extract and recognize the possible territory of traffic sign. The method of extracting the possible territory is to use the characteristics of YUV, YIQ, and CMYK color space from the RGB color space. Morphology uses the geometric characteristics of traffic sign to make the image segmentation. The recognition of traffic signs can be recognized by using the error back propagation algorithm. As a result of the experiment, the proposed system has proven its outstanding capability in extraction and recognition of candidate territory without the influence of differences in lighting and input image in various sizes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.7
• /
• pp.1468-1476
• /
• 1997

In mobile communication systems, the Reduced State Sequence Estimation(RSSE) receiver must be able to track changes in the channel. This is carried out by the adaptive channel estimator. However, when the tentative decisions are used in the channel estimator, incorrect decisions can cause error propagation. This paper presents a new channel estimator using the path history in the Viterbi decoder for preventing error propagation. The selection of the path history in the Viterbi decoder for preventing error propagation. The selection of the path history for the channel estimator depends on the path metric as in the decoding of the Viterbi decoder in RSSE. And a discussion on the channel estimator with different adaptation algorithms such as Least Mean Square(LMS) algorithm and Recursive Least Square(RLS) algorithm is provided. Results from computer simulations show that the RSSE receivers using the proposed channel estimator have better performance than the other conventional RSSE receiver, and that the channel estimator with RLS algorithm is adequate for multipath fading channel.

• Journal of KSNVE
• /
• v.4 no.1
• /
• pp.59-69
• /
• 1994

A new LMS algorithm titled constrained LMS' is proposed for problems with constrained structure. The conventional LMS algorithm can not be used because it destroys the constrained structures of the weights or parameters. Proposed method uses error-back propagation, which is popular in training neural networks, for error minimization. The illustrative examplesare shown to demonstrate the applicability of the proposed algorithm.