56016 research outputs found
Sort by
Direct aerosol optical depth retrievals using MODIS reflectance data and machine learning over East Asia
No full textAnthropogenic aerosols have rapidly increased since the industrial revolution and are harmful to human health. Moderate Resolution Imaging Spectroradiometer (MODIS) data are critical for retrieving aerosol properties worldwide. However, current MODIS aerosol optical depth (AOD) products require extensive computations and a precalculated lookup table. This study proposes assumption-free high-resolution AOD retrieval models based on the light gradient boosting machine method using MODIS data and ground-based observations over East Asia. The models were developed with three spatial resolutions: 250 m, 500 m, and 1 km. The results showed that 77.8% of the 250 m AOD values were within the MODIS expected error (EE) range, while 76.5%, 76.3%, and 70.08% of the 500 m, 1 km, and Multi-Angle Implementation of Atmospheric Correction (MAIAC) AOD values were within the EE range, respectively. Furthermore, an analysis of the time series and detailed spatial distribution of the proposed model-derived AOD based on data from the Korea???United States Air Quality campaign demonstrated the excellent quality of the 250 m AOD via further validation using a spatially independent dataset. The Shapley Additive exPlanations analysis identified the sensor zenith angle and top-of-atmosphere reflectance of the blue band as the key contributors to the models. In addition, while MAIAC has limited spatial coverage, the spatial frequency of the proposed direct AOD retrieval was nearly 1.5-times higher than that of the MAIAC AOD. Our findings confirmed that machine learning-based high-resolution AOD estimates can be obtained using only satellite data
Recent status of pre-bunching and re-bunching systems for low-energy experimental facilities at RAON
No full textRare isotope Accelerator complex for ON-line experiments (RAON), which is being constructed in Korea, will provide both rare isotope and stable ion beams for various nuclear physics experiments as well as other applications, at a wide energy range up to a few hundreds of MeV/nucleon. The ion beams will be provided to low-energy experimental facilities at RAON such as Korea Broad Acceptance Recoil spectrometer and Apparatus (KoBRA) and Nuclear Data Production System (NDPS). Since the bunch length of some ion beams becomes unsuitably long at the targets of these facilities, a re-bunching system has been designed and developed. Besides, a pre-bunching system based on a fast chopper and a Double Gap Buncher (DGB) has been developed and installed upstream of the Radio-Frequency Quadrupole (RFQ) to lower the repetition rate for the time-of-flight measurements of secondary particles at KoBRA and NDPS. We present the status of the pre-bunching and re-bunching systems for the RAON low-energy experimental facilities
Robustness in Portfolio Optimization
No full textPortfolio optimization is the basic quantitative approach for finding optimal portfolio weights. It has become increasingly important as portfolio construction involves more and more data and automated approaches. The inherent uncertainty in financial markets has led to consistent demand for improved robustness of portfolio models. In this article, the authors discuss the importance of robustness in portfolio optimization and present powerful methods that include robust estimators, robust portfolio optimization, distributionally robust optimization, and scenario-based optimization. They also review data-driven methods, machine learning???based models, and practical approaches for improving portfolio robustness
Deep Learning-Based Human Detection Using RGB and IR Images from Drones
No full textThis paper proposes an object detection algorithm based on a deep neural network which utilizes RGB and infrared (IR) images for human detection with drones. Although there are some public RGB image datasets in aerial view, there is no publicly available dataset in which both RGB and IR images are taken simultaneously with drones. Thus, we collect RGB and IR images at various altitudes on our own. However, the detection performance of the RGB and IR-based algorithm is limited, because there are more available RGB images for training than IR images or RGB and IR images taken simultaneously. To address this data imbalance, we use a generative adversarial neural network model called CycleGAN to generate IR images from RGB images in the public and self-collected dataset. Furthermore, the neural network for object detection is accelerated by neural network quantization and optimization to execute the algorithm faster on an embedded computing board. To shorten the search time and improve the robustness of detection, we employ a formation flight of multiple drones for human detection. The effectiveness of the integrated system of formation flight and onboard human detection algorithm is validated by the real flight experiment to find humans who are in a wide and wild area
Thermal conductivity controlled by a segregated network prepared using carbon nanotube/polyamide 6 composite with glass bubbles
No full textIn this study, a carbon nanotube-glass bubble/polyamide 6 (CNT-GB/PA6) multiscale hybrid composite was manufactured. Through a coagulation process including CNT and GB, a segregated network was formed to produce a composite structure with tunable thermal conductivity. Within the segregated network structure, a complex phenomenon of decrease and increase in thermal conductivity owing to the interaction by CNT and GB, and an equation to predict thermal conductivity through the contents of GB and CNT was formulated. A model to predict the thermal conductivity using the RSM analysis was presented, and the contents of GB and CNT were adjusted according to the required thermal conductivity. It was confirmed through the LFA thermal conductivity measurement that the thermal conductivity increased with GB and CNT contents. Moreover, when 30% of GB was added to the 5 wt% CNT composite, the thermal conductivity increased by about 17%. However, in the experiment to confirm the effect of GB alone, it was confirmed that the thermal conductivity decreased as the GB content increased. Thus, the size of the structural path, which was controlled by the GB content through which electrons pass, played an important role in this study. The results of this study can be used in astronautic fields that require insulation to save energy and in construction engineering where thermal insulation and heat emission are important
Exploring the Nonlinear Relationship between Development Density and Housing Price: Focusing on eXplainable AI (XAI)
No full text???????????? ???????????? ????????? ???????????? ?????? ??????, ?????? ??? ?????????????????? ???????????? ?????? ????????? ????????? ??????????????? ??????????????? ??????????????? ????????? ????????? ?????????????????? ????????????. ?????? ??????????????? ??????????????? ???????????? ?????? ????????? ????????? ?????? ?????????, ??????????????? ???????????? ??????????????? ????????? ???????????? ???????????? ??????????????? ????????? ???????????? ??????????????? ???????????????. ?????? 2000?????? ??????????????? ?????????????????? ????????? ????????? ???????????? ????????????, 2010?????? ?????? ????????? ?????????????????? ?????? ?????????????????? ?????? ????????? ????????? ??????????????? ???????????? ??????, ?????? ?????????????????? ?????? ???????????? ?????? ?????? ???????????? ??????. ????????? ??? ????????? ???????????? ???????????? ??????????????? ????????? ????????? ????????? ???????????? ?????? 2019??? 1????????? 2021??? 12????????? ????????? ????????? ??????????????? ????????? ???????????? ???????????? ?????? ????????? ?????? ????????? ????????????, ?????? ??????(??????, ????????????)??? ?????? ??????(?????????)??? ???????????? ????????? ???????????????. ??????, ???????????? ????????? ????????? ???????????? ????????? ???????????? ????????? ???????????? ????????? ?????? CatBoost ??????????????? ?????? ????????? ???????????? ?????? ??? ????????? SHAP??? ???????????? ???????????? ????????? ???????????? ?????? ?????? ???????????? ???????????? ?????? ???????????????. ???????????? ?????? ???????????? ???????????? ?????????????????? ?????? ????????? ??? 290~310%??? 490~510% ?????? ???????????? ???????????? ????????? ????????? ????????? ???????????? ????????? ????????? ????????? ????????????. ????????? ???????????? ????????? ????????? ?????? ?????? ?????? ?????? ??????????????? ??????, ?????? ??????????????? ????????? ???????????? ???????????? ?????? ???????????? ??? 900%??? ???????????? ???????????? ????????? ????????? ???????????? ?????? ??? ????????? ????????? ??? ?????????. ??? ????????? ????????? ??????????????? ????????? ?????? ?????? ?????? ????????? ????????????????????? ???????????? ???????????? ???????????? ???????????? ?????? ??????????????? ??? ??? ??????
Disentangling the mechanisms of equatorial Pacific climate change
No full textMost state-of-art models project a reduced equatorial Pacific east-west temperature gradient and a weakened Walker circulation under global warming. However, the causes of this robust projection remain elusive. Here, we devise a series of slab ocean model experiments to diagnostically decompose the global warming response into the contributions from the direct carbon dioxide (CO2) forcing, sea ice changes, and regional ocean heat uptake. The CO2 forcing dominates the Walker circulation slowdown through enhancing the tropical tropospheric stability. Antarctic sea ice changes and local ocean heat release are the dominant drivers for reduced zonal temperature gradient over the equatorial Pacific, while the Southern Ocean heat uptake opposes this change. Corroborating our model experiments, multimodel analysis shows that the models with greater Southern Ocean heat uptake exhibit less reduction in the temperature gradient and less weakening of the Walker circulation. Therefore, constraining the tropical Pacific projection requires a better insight into Southern Ocean processes
