• Journal of Integrative Natural Science
• /
• v.11 no.4
• /
• pp.204-208
• /
• 2018

Comparison of different optimizer performance in photovoltaic power modeling using artificial neural deep learning techniques is described in this paper. Six different deep learning optimizers are tested for Long-Short-Term Memory networks in this study. The optimizers are namely Adam, Stochastic Gradient Descent, Root Mean Square Propagation, Adaptive Gradient, and some variants such as Adamax and Nadam. For comparing the optimization techniques, high and low fluctuated photovoltaic power output are examined and the power output is real data obtained from the site at Mokpo university. Using Python Keras version, we have developed the prediction program for the performance evaluation of the optimizations. The prediction error results of each optimizer in both high and low power cases shows that the Adam has better performance compared to the other optimizers.

• Journal of Korean Philosophical Society
• /
• v.142
• /
• pp.27-50
• /
• 2017

When we lose our memories or when our self-identity becomes blurred, we may feel as if our mind has split into multiple minds. If someone makes another 'me' by copying my body and mind, how should two conscious 'I's change their credences? In this article, we present a new thought experiment that can be called the 'Two Adams Problem'. This thought experiment represents a piece of philosophical inquiry that speculates on the nature of self-consciousness information that 'I am awake now'. On Sunday evening, as soon as Adam falls asleep, the philosopher Roro makes another Adam by copying Adam's original body and mind. They sleep separately in another room. Within a minute, Roro throws a fair coin. If the coin lands heads up Roro wakes just one of the Adams up on Monday. If the coin lands tails up Roro wakes both Adams up. On Monday, at least one of the Adams will wake up. To what degree ought they believe that the outcome of the coin toss is heads? We will argue that the correct answer to this question is 1/2.

• Communications of the Korean Institute of Information Scientists and Engineers
• /
• v.26 no.10
• /
• pp.28-40
• /
• 2008

• 대한생식의학회:학술대회논문집
• /
• 2002.11a
• /
• pp.85-86
• /
• 2002

• 대한생식의학회:학술대회논문집
• /
• 2002.11a
• /
• pp.86-86
• /
• 2002

• Clinical Endoscopy
• /
• v.57 no.5
• /
• pp.694-696
• /
• 2024

• ALGAE
• /
• v.37 no.4
• /
• pp.301-316
• /
• 2022

The red alga Compsopogon caeruleus can generally be found in tropical and subtropical waters worldwide. In addition to its natural habitats, this species may be found in waters that receive abnormally hot water, e.g., from powerhouses. To date, the presence of C. caeruleus has not been observed in thermally polluted lacustrine ecosystems in Poland, which has a moderate climate. The thalli of this red alga were found growing on Vallisneria spiralis in Lichenskie Lake. Importantly, this paper presents a previously unknown relationship between the temperature (20, 25, 30, 35, and 40℃) and oxygen requirements of C. caeruleus (based on ex situ measurements of O₂ consumption by thalli). Surprisingly, 35℃ can be the optimum temperature for C. caeruleus, and this temperature is higher than the values reported by some previous thermal analyses by approximately 10℃. Additionally, we reviewed and mapped the distribution of this nonnative and mesophilic red alga in natural / seminatural water ecosystems in Europe. Finally, we propose that the occurrence of C. caeruleus mature thalli can be a novel, simple and easy-to-recognize bioindicator of artificially and permanently heated waters in moderate climate zones by a regular discharge of postindustrial water.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1329-1341
• /
• 2022

Destroying the marine environment and marine ecosystem and causing marine accidents, marine debris is generated every year, and among them, submerged marine debris is difficult to identify and collect because it is on the seabed. Therefore, deep-learning-based semantic segmentation was experimented on waste fish nets and waste ropes using underwater images to identify efficient collection and distribution. For segmentation, a high-resolution network (HRNet), a state-of-the-art deep learning technique, was used, and the performance of each optimizer was compared. In the segmentation result fish net, F1 score=(86.46%, 86.20%, 85.29%), IoU=(76.15%, 75.74%, 74.36%), For the rope F1 score=(80.49%, 80.48%, 77.86%), IoU=(67.35%, 67.33%, 63.75%) in the order of adaptive moment estimation (Adam), Momentum, and stochastic gradient descent (SGD). Adam's results were the highest in both fish net and rope. Through the research results, the evaluation of segmentation performance for each optimizer and the possibility of segmentation of marine debris in the latest deep learning technique were confirmed. Accordingly, it is judged that by applying the latest deep learning technique to the identification of submerged marine debris through underwater images, it will be helpful in estimating the distribution of marine sedimentation debris through more accurate and efficient identification than identification through the naked eye.

• Bulletin of the Korean Chemical Society
• /
• v.34 no.10
• /
• pp.3147-3149
• /
• 2013

• Atmosphere
• /
• v.26 no.2
• /
• pp.257-275
• /
• 2016

Asian dust is a seasonal meteorological phenomenon influencing most East Asia, irregularly occurring during spring. Unusual heavy Asian dust event in winter was observed in Seoul, Korea, with up to $1,044{\mu}g\;m^{-3}$ of hourly mean $PM_{10}$, in 22~23 February 2015. Causes of such infrequent event has been studied using both ground based and spaceborne observations, as well as numerical simulations including ECMWF ERA Interim reanalysis, NOAA HYSPLIT backward trajectory analysis, and ADAM2-Haze simulation. Analysis showed that southern Mongolia and northern China, one of the areas for dust origins, had been warm and dry condition, i.e. no snow depth, soil temperature of ${\sim}0^{\circ}C$, and cumulative rainfall of 1 mm in February, along with strong surface winds higher than critical wind speed of $6{\sim}7.5m\;s^{-1}$ during 20~21 February. While Jurihe, China, ($42^{\circ}23^{\prime}56^{{\prime}{\prime}}N$, $112^{\circ}53^{\prime}58^{{\prime}{\prime}}E$) experienced $9,308{\mu}g\;m^{-3}$ of hourly mean surface $PM_{10}$ during the period, the Asian dust had affected the Korean Peninsula within 24 hours traveling through strong north-westerly wind at ~2 km altitude. KMA issued Asian dust alert from 1100 KST on 22nd to 2200 KST on 23rd since above $400{\mu}g\;m^{-3}$ of hourly mean surface $PM_{10}$. It is also important to note that, previously to arrival of the Asian dust, the Korean Peninsula was affected by anthropogenic air pollutants ($NO_3^-$, $SO_4^{2-}$, and $NH_4^+$) originated from the megacities and large industrial areas in northeast China. In addition, this study suggests using various data sets from modeling and observations as well as improving predictability of the ADAM2-Haze model itself, in order to more accurately predict the occurrence and impacts of the Asian dust over the Korean peninsula.