• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.344-355
• /
• 2002

Having a variety of good characteristics against other pattern recognition techniques, Multilayer Perceptron (MLP) has been used in wide applications. But, it is known that Error Backpropagation (EBP) algorithm which MLP uses in learning has a defect that requires relatively long leaning time. Because learning data in pattern recognition contain abundant redundancies, in order to increase learning speed it is very effective to use online-based teaming methods, which update parameters of MLP pattern by pattern. Typical online EBP algorithm applies fixed learning rate for each update of parameters. Though a large amount of speedup with online EBP can be obtained by choosing an appropriate fixed rate, fixing the rate leads to the problem that the algorithm cannot respond effectively to different leaning phases as the phases change and the learning pattern areas vary. To solve this problem, this paper defines learning as three phases and proposes a Instant Learning by Varying Rate and Skipping (ILVRS) method to reflect only necessary patterns when learning phases change. The basic concept of ILVRS is as follows. To discriminate and use necessary patterns which change as learning proceeds, (1) ILVRS uses a variable learning rate which is an error calculated from each pattern and is suppressed within a proper range, and (2) ILVRS bypasses unnecessary patterns in loaming phases. In this paper, an experimentation is conducted for speaker verification as an application of pattern recognition, and the results are presented to verify the performance of ILVRS.

• Applied Chemistry for Engineering
• /
• v.7 no.5
• /
• pp.877-887
• /
• 1996

The ideal membranes for membrane-covered oxygen probes system should be selectively permeable for oxygen and chemically inert, and have good mechanical strength. Polysulfone(PSf) was selected to develop the membrane for membrane-covered oxygen electrodes system. PSf membranes have properties such as good reproducibility, good mechanical strength, chemical inertness, and high heat resistance. PSf membranes were cast from polymer solution on the glass plate at constant temperature, and casting solvents used were tetrahydrofuran(THF), methylene chloride, and N-methyl-2-pyrrolidone(NMP). Tricresyl phosphate(TCP) as plasicizer was added to PSf to increase the softness of membrane. The permeation characteristics were observed for pure oxygen and nitrogen through pure PSf membranes by variable volume method and membrane-covered electrode system. The permeability coefficients of oxygen and nitrogen measured by variable volume method were slightly decreased with increasing of upstream pressure. The permeation properties of PSf membrane using methylene choride as casting solvent were not affected by the PSf amount of polymer solution. The permeability coefficients of oxygen and nitrogen for PSf membrane containing TCP were very slightly lower than those for pure PSf membrane, but ideal separation factors were slightly higher. The flexibility of PSf membrane containing 2wt% TCP was better than that of pure PSf membrane. It was expected that this increase in flexibility would solve the difficulty of fixing the membrane to the cathode. The membrane-covered oxygen probes system was composed of anode, cathode and electrolyte. The type of the anode was Ag/AgCl half-cell, that of cathode was Ag, and the electrolyte was 4N KCl solution. The result of sampled current voltametry for PSf membrane showed the plateu region at -0.3V~-1.0V. The correlation coefficient of oxygen partial pressure versus current for PSf membrane was relatively high, 0.99949. It was concluded that PSf membrane was the good candidate for the membrane-covered oxygen probes system.

• The Journal of the Korea Contents Association
• /
• v.8 no.11
• /
• pp.210-216
• /
• 2008

In this study, image comparisons were carried out using a MRI contrast medium which was derived by mixing a polyaminocarboxylic ligand and a gadolinium (III) transition metal which is paramagnetic and has good neutron absorbing capabilities with Gd-DTPA which is currently being used widely in the clinical setting. By using a 1.0T (Harmony, SIEMENS) MR equipment, phantoms of which 100cc of saline was diluted with a diethylenetriaminetriacetic acid derivative and Gd-DTPA were imaged. The amount of diethylenetriaminetriacetic acid and Gd-DTPA which was diluted into the 100cc of saline was 0.05mmol/L, 0.1mmol/L, 0.15mmol/L, 0.2mmol/L, 0.3mmol/L, 0.5mmol/L, 1.0mmol/L, 2.0mmol/L, 3.0mmol/L and 4.9mmol/L respectively. Head coils were used and while fixing the SE pulse sequence and image variable (as TE is 14ms, 1NEX with a 256x201 matrix), the signal intensity and simple contrast ratios according to changing concentrations and TR were compared with various TR at 300ms, 400ms, 500ms, 600ms, 700ms, 800ms, 900ms, 1000ms, 1200ms, 1400ms and 1600ms. According to the comparison results of the signal intensity of the image based on changes in contrast medium concentrations and TR, the differences in signal intensity between the two contrast mediums were found to be small at $1.0{\sim}2.0mmol/L$ when the highest signal intensity was achieved. However, at concentrations of 1.0mmol/L or less, the signal intensity was markedly higher in the Diethylenetriaminetriacetic acid derivative than in the Gd-DTPA complex. It was also found that the differences in the signal intensities demonstrated by the concentrations of the contrast mediums were affected by the TR. Accordingly, the efficacy of the Diethylenetriaminetriacetic acid derivative was shown to be better than the Gd-DTPA and also possible to get the optimum image quality by the use of an appropriate TR with appropriate concentrations of contrast medium.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1041-1043
• /
• 1993

Fuzzy logic based Control Theory has gained much interest in the industrial world, thanks to its ability to formalize and solve in a very natural way many problems that are very difficult to quantify at an analytical level. This paper shows a solution for treating membership function inside hardware circuits. The proposed hardware structure optimizes the memoried size by using particular form of the vectorial representation. The process of memorizing fuzzy sets, i.e. their membership function, has always been one of the more problematic issues for the hardware implementation, due to the quite large memory space that is needed. To simplify such an implementation, it is commonly [1,2,8,9,10,11] used to limit the membership functions either to those having triangular or trapezoidal shape, or pre-definite shape. These kinds of functions are able to cover a large spectrum of applications with a limited usage of memory, since they can be memorized by specifying very few parameters ( ight, base, critical points, etc.). This however results in a loss of computational power due to computation on the medium points. A solution to this problem is obtained by discretizing the universe of discourse U, i.e. by fixing a finite number of points and memorizing the value of the membership functions on such points [3,10,14,15]. Such a solution provides a satisfying computational speed, a very high precision of definitions and gives the users the opportunity to choose membership functions of any shape. However, a significant memory waste can as well be registered. It is indeed possible that for each of the given fuzzy sets many elements of the universe of discourse have a membership value equal to zero. It has also been noticed that almost in all cases common points among fuzzy sets, i.e. points with non null membership values are very few. More specifically, in many applications, for each element u of U, there exists at most three fuzzy sets for which the membership value is ot null [3,5,6,7,12,13]. Our proposal is based on such hypotheses. Moreover, we use a technique that even though it does not restrict the shapes of membership functions, it reduces strongly the computational time for the membership values and optimizes the function memorization. In figure 1 it is represented a term set whose characteristics are common for fuzzy controllers and to which we will refer in the following. The above term set has a universe of discourse with 128 elements (so to have a good resolution), 8 fuzzy sets that describe the term set, 32 levels of discretization for the membership values. Clearly, the number of bits necessary for the given specifications are 5 for 32 truth levels, 3 for 8 membership functions and 7 for 128 levels of resolution. The memory depth is given by the dimension of the universe of the discourse (128 in our case) and it will be represented by the memory rows. The length of a world of memory is defined by: Length = nem (dm(m)＋dm(fm) Where: fm is the maximum number of non null values in every element of the universe of the discourse, dm(m) is the dimension of the values of the membership function m, dm(fm) is the dimension of the word to represent the index of the highest membership function. In our case then Length=24. The memory dimension is therefore 128*24 bits. If we had chosen to memorize all values of the membership functions we would have needed to memorize on each memory row the membership value of each element. Fuzzy sets word dimension is 8*5 bits. Therefore, the dimension of the memory would have been 128*40 bits. Coherently with our hypothesis, in fig. 1 each element of universe of the discourse has a non null membership value on at most three fuzzy sets. Focusing on the elements 32,64,96 of the universe of discourse, they will be memorized as follows: The computation of the rule weights is done by comparing those bits that represent the index of the membership function, with the word of the program memor . The output bus of the Program Memory (μCOD), is given as input a comparator (Combinatory Net). If the index is equal to the bus value then one of the non null weight derives from the rule and it is produced as output, otherwise the output is zero (fig. 2). It is clear, that the memory dimension of the antecedent is in this way reduced since only non null values are memorized. Moreover, the time performance of the system is equivalent to the performance of a system using vectorial memorization of all weights. The dimensioning of the word is influenced by some parameters of the input variable. The most important parameter is the maximum number membership functions (nfm) having a non null value in each element of the universe of discourse. From our study in the field of fuzzy system, we see that typically nfm 3 and there are at most 16 membership function. At any rate, such a value can be increased up to the physical dimensional limit of the antecedent memory. A less important role n the optimization process of the word dimension is played by the number of membership functions defined for each linguistic term. The table below shows the request word dimension as a function of such parameters and compares our proposed method with the method of vectorial memorization[10]. Summing up, the characteristics of our method are: Users are not restricted to membership functions with specific shapes. The number of the fuzzy sets and the resolution of the vertical axis have a very small influence in increasing memory space. Weight computations are done by combinatorial network and therefore the time performance of the system is equivalent to the one of the vectorial method. The number of non null membership values on any element of the universe of discourse is limited. Such a constraint is usually non very restrictive since many controllers obtain a good precision with only three non null weights. The method here briefly described has been adopted by our group in the design of an optimized version of the coprocessor described in [10].