• Electronics and Telecommunications Trends
• /
• v.38 no.4
• /
• pp.95-103
• /
• 2023

The importance of the decentralized training with decentralized execution (DTDE) framework is well-known in the study of multiagent reinforcement learning. In many real-world environments, agents cannot share information. Hence, they must be trained in a decentralized manner. However, the DTDE framework has been less studied than the centralized training with decentralized execution framework. One of the main reasons is that many problems arise when training agents in a decentralized manner. For example, DTDE algorithms are often computationally demanding or can encounter problems with non-stationarity. Another reason is the lack of simulation environments that can properly handle the DTDE framework. We discuss current research trends in the DTDE framework.

• Journal of the Korean Mathematical Society
• /
• v.61 no.4
• /
• pp.713-742
• /
• 2024

From the ordinary notion of upper-tail quantitle function, a new concept called conditionally upper-tail quantitle function given a σ-algebra is proposed. Some basic properties of this terminology and further properties of conditionally strictly stationary sequences are derived. By means of these properties, several conditional central limit theorems for a sequence of conditionally strong mixing and conditionally strictly stationary random variables are established, some of which are the conditional versions corresponding to earlier results under non-conditional case.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.152-152
• /
• 2016

최근 기후변화의 영향으로 장마, 태풍 등 극한사상의 발생빈도와 강도가 비정상적인 증가 추세를 나타내고 있으며, 여름철 국지성 호우로 인한 농경지 및 도심 저지대 지역의 침수 피해가 발생하고 있다. 침수 피해에 대한 대책 마련을 위해서는 수공구조물 설계 기준을 초과하는 호우에 대한 홍수 영향을 분석할 필요가 있으며, 기후변화에 따른 강우자료의 변화 특성을 파악하기 위해서는 비정상성 (Non-Stationary) 가정이 수반되어야 한다. 따라서 본 연구에서는 비정상성 빈도해석을 통해 중소하천을 대상으로 부정류 해석을 실시하고 미래 침수특성을 분석하고자 한다. 연구대상지는 상습 침수지역이 위치한 중소하천을 선정하였고, 각 유역에 가장 인접한 기상관측소로부터 강수량 자료를 수집하였다. 강수량 모의 자료는 국립기상과학원에서 제공하는 해상도 12.5 km의 지역 기후변화 시나리오를 이용하여 구축하였다. 구축한 강수량 자료는 정상성 및 비정상성 빈도해석을 각각 수행하였으며 비정상성 빈도해석 방법으로는 누적평균 방법 및 이동평균 방법을 적용하였다. 유역 유출량은 실무에서 설계홍수량 산정에 널리 이용되고 있는 HEC-HMS 모형으로 산정하였다. 유출량과 하천기본계획의 하천단면 측량자료를 1차원 부정류 해석 모형인 HEC-RAS 모형에 입력하고 부정류 해석을 실시하여 하천 홍수위를 모의하였다. 본 연구의 결과는 상습 침수 지역의 침수 피해에 대한 관리 대책을 수립하는데 기초자료로 사용할 수 있을 것을 사료된다.

• Journal of Communications and Networks
• /
• v.9 no.1
• /
• pp.18-27
• /
• 2007

The conventional normalized least mean square (NLMS) algorithm is the most widely used for adaptive identification within a non-stationary input context. The convergence of the NLMS algorithm is independent of environmental changes. However, its steady state performance is impaired during input sequences with low dynamics. In this paper, we propose a new NLMS algorithm which is, in the steady state, insensitive to the time variations of the input dynamics. The square soot (SR)-NLMS algorithm is based on a normalization of the LMS adaptive filter input by the Euclidean norm of the tap-input. The tap-input power of the SR-NLMS adaptive filter is then equal to one even during sequences with low dynamics. Therefore, the amplification of the observation noise power by the tap-input power is cancelled in the misadjustment time evolution. The harmful effect of the low dynamics input sequences, on the steady state performance of the LMS adaptive filter are then reduced. In addition, the square root normalized input is more stationary than the base input. Therefore, the robustness of LMS adaptive filter with respect to the input non stationarity is enhanced. A performance analysis of the first- and the second-order statistic behavior of the proposed SR-NLMS adaptive filter is carried out. In particular, an analytical expression of the step size ensuring stability and mean convergence is derived. In addition, the results of an experimental study demonstrating the good performance of the SR-NLMS algorithm are given. A comparison of these results with those obtained from a standard NLMS algorithm, is performed. It is shown that, within a non-stationary input context, the SR-NLMS algorithm exhibits better performance than the NLMS algorithm.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.20 no.3
• /
• pp.125-140
• /
• 2017

Until recently, regression based spatial interpolation methods and Kriging based spatial interpolation methods have been largely used to estimate land price or housing price, but less attention has been paid on comparing the performance of these spatial interpolation methods. In this regard, this research applied regression based spatial interpolators and Kriging based spatial interpolators for estimating the land prices in Dalseo-gu, Daegu metropolitan city and evaluated the accuracy of eight spatial interpolators. OLS, SLM, SEM, and GWR were used as regression based spatial interpolators while SK, OK, UK, and CK were employed as Kriging based spatial interpolators. The global accuracy was statistically evaluated by RMSE, adjusted RMSE, and COD. The relative accuracy was visually compared by three-dimensional residual error map and scatterplot. Results from statistical and visual analyses indicate that GWR reflecting the spatial non-stationarity was a relatively more accurate spatial predictor to estimate land prices in the study area than SAR and Kriging based spatial interpolators considering the spatial dependence. The findings from this research will contribute to the secondary research into analyzing the urban spatial structure with land prices.

• Korean Journal of Ecology and Environment
• /
• v.52 no.2
• /
• pp.94-104
• /
• 2019

In this study, we investigated the spatial variation in land use and topographic effects on water quality at the Geum river watershed in South Korea, using the ordinary least squares(OLS) and geographically weighted regression (GWR) models. Understanding the complex interactions between land use, slope, elevation, and water quality is essential for water pollution control and watershed management. We monitored four water quality indicators -total phosphorus, total nitrogen, biochemical oxygen demand, and dissolved oxygen levels - across three land use types (urban, agricultural, and forested) and two topographic features (elevation and mean slope). Results from GWR modeling revealed that land use and topography did not affect water quality consistently through space, but instead exhibited substantial spatial non-stationarity. The GWR model performed better than the OLS model as it produced a higher adjusted $R^2$ value. Spatial variation in interactions among variables could be visualized by mapping $R^2$ values from the GWR model at fine spatial resolution. Using the GWR model, we were able to identify local pollution sources, determine habitat status, and recommend appropriate land-use planning policies for watershed management.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.5
• /
• pp.222-228
• /
• 2010

In this paper, we propose a new approach to noise estimation incorporating spectral deviation with soft decision scheme to enhance the intelligibility of the degraded speech signal in non-stationary noisy environments. Since the conventional noise estimation technique based on soft decision scheme estimates and updates the noise power spectrum using a fixed smoothing parameter which was assumed in stationary noisy environments, it is difficult to obtain the robust estimates of noise power spectrum in non-stationary noisy environments that spectral characteristics of noise signal such as restaurant constantly change. In this paper, once we first classify the stationary noise and non-stationary noise environments based on the analysis of spectral deviation of noise signal, we adaptively estimate and update the noise power spectrum according to the classified noise types. The performances of the proposed algorithm are evaluated by ITU-T P. 862 perceptual evaluation of speech quality (PESQ) under various ambient noise environments and show better performances compared with the conventional method.

• Journal of Wetlands Research
• /
• v.20 no.4
• /
• pp.338-344
• /
• 2018

In this study, the surface air temperature (SAT) and the dew-point temperature (DPT) are applied as the covariance of the location parameter among three parameters of GEV distribution to reflect the non-stationarity of extreme rainfall due to climate change. Busan station is selected as the study site and the monthly maximum daily rainfall depth from May to October is used for analysis. Various models are constructed to select the most appropriate co-variate(SAT and DPT) function for location parameter of GEV distribution, and the model with the smallest AIC(Akaike Information Criterion) is selected as the optimal model. As a result, it is found that the non-stationary GEV distribution with co-variate of exp(DPT) is the best. The selected model is used to analyze the effect of climate change scenarios on extreme rainfall quantile. It is confirmed that the design rainfall depth is highly likely to increase as the future DPT increases.

• Journal of Korean Society on Water Environment
• /
• v.26 no.2
• /
• pp.252-260
• /
• 2010

The point rainfall measurements need to be converted to the areal rainfall by means of mean areal precipitation (MAP) estimation methods. And it is not appropriate to evaluate the areal rainfall with constant drift because of the geomorphological influences to rainfall field. Non-stationarity should be applied to the estimation of the areal rainfall, therefore, to consider these effects. Kriging methods with special functional would be a suitable tool in this case. Generalized covariance Kriging method is the most developed one among different Kriging methods. From this point of view this study performs the analysis of its applicability to distributed runoff model. For these purpose, distributed rainfall was created by Thiessen and Kriging method. And distributed rainfall of each method was applied into HyGIS-GRM. The result of applying, Runoff was different in the rainfall data form. Therefore, To apply Kriging method with physical meaning is that it is the useful method as distributed rainfall-runoff model.

• Journal of Ecology and Environment
• /
• v.43 no.2
• /
• pp.191-201
• /
• 2019

Macroecologists and biogeographers continue to predict the distribution of species across space based on the relationship between biotic processes and environmental variables. This approach uses data related to, for example, species abundance or presence/absence, climate, geomorphology, and soils. Researchers have acknowledged in their statistical analyses the importance of accounting for the effects of spatial autocorrelation (SAC), which indicates a degree of dependence between pairs of nearby observations. It has been agreed that residual spatial autocorrelation (rSAC) can have a substantial impact on modeling processes and inferences. However, more attention should be paid to the sources of rSAC and the degree to which rSAC becomes problematic. Here, we review previous studies to identify diverse factors that potentially induce the presence of rSAC in macroecological and biogeographical models. Furthermore, an emphasis is put on the quantification of rSAC by seeking to unveil the magnitude to which the presence of SAC in model residuals becomes detrimental to the modeling process. It turned out that five categories of factors can drive the presence of SAC in model residuals: ecological data and processes, scale and distance, missing variables, sampling design, and assumptions and methodological approaches. Additionally, we noted that more explicit and elaborated discussion of rSAC should be presented in species distribution modeling. Future investigations involving the quantification of rSAC are recommended in order to understand when rSAC can have an adverse effect on the modeling process.