• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.66-69
• /
• 2003

This paper considers a reinforcement learning(RL) which deals with real environments. Most reinforcement learning studies have been made by simulations because real-environment learning requires large computational cost and much time. Furthermore, it is more difficult to acquire many rewards efficiently in real environments than in virtual ones. The most important requirement to make real-environment learning successful is the appropriate construction of the state space. In this paper, to begin with, I show the basic overview of the reinforcement learning under real environments. Next, 1 introduce a state-space construction method under real environmental which is State Partition Method. Finally I apply this method to a robot navigation problem and compare it with conventional methods.

• Natural Product Sciences
• /
• v.15 no.3
• /
• pp.144-150
• /
• 2009

Gradient centrifugal partition chromatography (GCPC) method was developed and applied to isolate 3,5-dimethoxyphenanthrene-2,7-diol (DMP) and batatasin-I (BA-I) from the dichloromethane soluble extract of Dioscorea opposita. In this method, the lower phase of n-hexane-methanol-water system (HMW, 10 : 9 : 1, v/v) was used as a mobile phase A (MpA) and water was used as a mobile phase B (MpB). This gradient CPC method is comparable to that of reversed-phase HPLC method in that the stationary upper-phase of HMW (10 : 9 : 1 v/v) works as if it were reversed-phase silica gel due to its hydrophobic property, while the lower phase (MpA) and water (MpB) functioned as hydrophilic mobile phases. The initial condition of the mobile phase was 20% MpA/80% MpB and maintained for 150 min to obtain DMP (1.2 mg), and then MpA was increased up to 50% to elute BA-I (1.7 mg). The purities of DMP and BA-I were 94.1% and 98.3% with the recovery yields of 83% and 86%, respectively. Similar results were obtained by linear-gradient CPC. The CPC peak fractions were identified by comparing their retention time to those of authentic samples of DMP and BA-I and their spectroscopic data ($^1$H NMR and $^{13}$C NMR) to those of literature values.

• Korean Journal of Weed Science
• /
• v.17 no.3
• /
• pp.303-309
• /
• 1997

Variation of morphological and physiological traits of 50 Bidens tripartita accessions were studied and the accessions were grouped through cluster analysis based on four major characters; plant type, leaf partition, achene length, days to flowering. Bidens tripartite accessions have shown significant variations in plant type, stem length, days to flowering, leaf shape, leaf partition, chlorophyll content, leaf color, stem color, achene color, achene length and achene shape. Most of Bidens tripartite accessions appeared to have strong dormancy and also photodormancy with some exceptions. Plants could be classified into 5 types from straight(I) to triangle(V), and intermediate diamond type(III) was prevalent. The plant type score has negative correlation with the stem length. None, three, and five part leaved plants were observed and most of them were three or five parted. Leaf partition had negative correlation with achene length and chlorophyll content. Average days to flowering was 108 days in the range of 94~141 days. It had positive correlation with achene length and leaf shape and negative correlation with achene color. Average achene length was 10.0mm and it had positive correlation with achene shape, stem length, days to flowering and leaf shape. It also had negative correlation with leaf color, stem color, achene color, leaf partition. Bidens tripartite accessions could be divided into identifiable six groups from the cluster analysis at the distance 0.06 using Ward's minimum-variance method.

• Analytical Science and Technology
• /
• v.6 no.1
• /
• pp.1-8
• /
• 1993

The purpose of this study is to investigate the elution behavior of monosubstituted phenols and benzenes in micellar liquid chromatographic system, $C_{18}$ column-anionic surfactant, sodium dodecyl sulfate(SDS). The partition coefficients between the micellar pseudophase-water and modified stationary phase-water are calculated by the relationship between solute retention and micellar mobile phase(SDS) composition. The free energy of transfer of solute from water to micelle is also calculated from these values. There is a direct correlation between the hydrophobicity parameters in MLC and corresponding partition data for 1-octanol-water, which indicates that the hydrophobicity of molecules plays an important role in the partition for both systems and that quantitative structure activity relationships(QSAR) are available from studies on micellar partition. The other purpose of this study is to investigate methylene selectivity of alkyl homologous series through correlation between retention and the number of carbons. The correlation between hydrophobicity parameters in MLC and 1-octanol-water partition data was also observed when n-propanol was as a modifier in the mobile phase.

• Nuclear Engineering and Technology
• /
• v.31 no.6
• /
• pp.572-585
• /
• 1999

The characteristics of two-phase flow instability in a vertical boiling channel connecting with an unheated riser are investigated through the linear stability analysis model. Various two-phase flow models, including thermal non-equilibrium effects, are taken into account for establishing a physical model in the time domain. A classical approach to the frequency response method is adopted for the stability analysis by employing the D-partition method. The adequacy of the linear model is verified by evaluating experimental data at high quality conditions. It reveals that the flow-pattern-dependent drift velocity model enhances the prediction accuracy while the homogeneous equilibrium model shows the most conservative predictions. The characteristics of density wave oscillations under low-power and low-quality conditions are investigated by devising a simple model which accounts for the gravitational and frictional pressure losses along the channel. The necessary conditions for the occurrences of type-I instability and flow excursion are deduced from the one-dimensional D-partition analysis. The parametric effects of some design variables on low quality oscillations are also investigated.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.9 no.3
• /
• pp.312-322
• /
• 1997

The natural convective flow in a two-dimensional square enclosure with horizontal partitions is investigated numerically. The enclosure was composed of the lower hot and the upper cold horizontal walls and the adiabatic vertical walls, and two identical partitions were positioned perpendicularly at the mid-height of the right and left walls, respectively. The governing equations are solved by using the finite element method with Galerkin method. Calculations are made for different partition lengths, partition conductivites, and Rayleigh numbers based on the temperature difference between two horizontal walls and the enclosure height with water(Pr=4.95). An oscillatory motion of the natural convective flow is affected significantly by the variation of the gap width and Rayleigh number. When the gap width is comparatively short, the heat transfer rate is raised with the increase of the thermal conductivity of partitions. However, for sufficiently large gap widths at higher Rayleigh numbers, the average Nusselt numbers of the conductive partitions are smaller than those of the adiabatic partitions.

• Environmental Engineering Research
• /
• v.11 no.1
• /
• pp.33-44
• /
• 2006

The objective of this study was to develop and assess a method for estimating the rain scavenging ratios (RSRs) of particle-bound PAHs and PCBs using measured scavenging ratio of particulate matters (PM) and routinely available data of physico-chemical properties of PM. Paired atmospheric and rainwater sampling was conducted for a total of 4 rain events. Assuming equilibrium partitioning in rainwater-gas-PM system, an equation was derived for estimating the RSR of particle-bound chemicals as a function of RSR of PM and three equilibrium partition constants (i.e. dimensionless Henry's law constant, gas-particle partition coefficient, and water-particle partition coefficient). For all PAHs, the model significantly under-predicted the RSR while the model prediction for PCBs agreed with observation mostly within a factor of 5. The RSR model for the chemicals is of limited use as its accuracy critically depends on how close the observed partitioning of the chemicals in the gas-PM-rainwater system is to that estimated under the equilibrium assumption.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.11
• /
• pp.69-75
• /
• 2015

This paper suggests heuristic polynomial time algorithm for crew scheduling problem that is a kind of optimization problems. This problem has been solved by linear programming, set cover problem, set partition problem, column generation, etc. But the optimal solution has not been obtained by these methods. This paper sorts transit costs $c_{ij}$ to ascending order, and the task i and j crew paths are merged in case of the sum of operation time ${\Sigma}o$ is less than day working time T. As a result, we can be obtain the minimum number of crews $_{min}K$ and minimum transit cost $z=_{min}c_{ij}$. For the transit cost of specific number of crews $K(K>_{min}K)$, we delete the maximum $c_{ij}$ as much as the number of $K-_{min}K$, and to partition a crew path. For the 5 benchmark data, this algorithm can be gets less transit cost than state-of-the-art algorithms, and gets the minimum number of crews.

• Journal of the Korean Mathematical Society
• /
• v.45 no.6
• /
• pp.1613-1622
• /
• 2008

Let G be a graph with vertex set V(G) and edge set E(G), and let g, f be two nonnegative integer-valued functions defined on V(G) such that $g(x)\;{\leq}\;f(x)$ for every vertex x of V(G). We use $d_G(x)$ to denote the degree of a vertex x of G. A (g, f)-factor of G is a spanning subgraph F of G such that $g(x)\;{\leq}\;d_F(x)\;{\leq}\;f(x)$ for every vertex x of V(F). In particular, G is called a (g, f)-graph if G itself is a (g, f)-factor. A (g, f)-factorization of G is a partition of E(G) into edge-disjoint (g, f)-factors. Let F = {$F_1$, $F_2$, ..., $F_m$} be a factorization of G and H be a subgraph of G with mr edges. If $F_i$, $1\;{\leq}\;i\;{\leq}\;m$, has exactly r edges in common with H, we say that F is r-orthogonal to H. If for any partition {$A_1$, $A_2$, ..., $A_m$} of E(H) with $|A_i|=r$ there is a (g, f)-factorization F = {$F_1$, $F_2$, ..., $F_m$} of G such that $A_i\;{\subseteq}E(F_i)$, $1\;{\leq}\;i\;{\leq}\;m$, then we say that G has (g, f)-factorizations randomly r-orthogonal to H. In this paper it is proved that every (0, mf - (m - 1)r)-graph has (0, f)-factorizations randomly r-orthogonal to any given subgraph with mr edges if $f(x)\;{\geq}\;3r\;-\;1$ for any $x\;{\in}\;V(G)$.

• International Journal of Reliability and Applications
• /
• v.12 no.2
• /
• pp.103-116
• /
• 2011

A series-parallel system with independent but non-identical components is considered. The expressions have been derived for the joint reliability importance (JRI) of m (${\geq}2$) components, chosen from a series-parallel system. JRIs of components of two different series-parallel systems are studied analytically and graphically.