• Journal of Korean Association for Spatial Structures
• /
• v.23 no.2
• /
• pp.29-36
• /
• 2023

Reinforcement learning (RL) is widely applied to various engineering fields. Especially, RL has shown successful performance for control problems, such as vehicles, robotics, and active structural control system. However, little research on application of RL to optimal structural design has conducted to date. In this study, the possibility of application of RL to structural design of reinforced concrete (RC) beam was investigated. The example of RC beam structural design problem introduced in previous study was used for comparative study. Deep q-network (DQN) is a famous RL algorithm presenting good performance in the discrete action space and thus it was used in this study. The action of DQN agent is required to represent design variables of RC beam. However, the number of design variables of RC beam is too many to represent by the action of conventional DQN. To solve this problem, multi-agent DQN was used in this study. For more effective reinforcement learning process, DDQN (Double Q-Learning) that is an advanced version of a conventional DQN was employed. The multi-agent of DDQN was trained for optimal structural design of RC beam to satisfy American Concrete Institute (318) without any hand-labeled dataset. Five agents of DDQN provides actions for beam with, beam depth, main rebar size, number of main rebar, and shear stirrup size, respectively. Five agents of DDQN were trained for 10,000 episodes and the performance of the multi-agent of DDQN was evaluated with 100 test design cases. This study shows that the multi-agent DDQN algorithm can provide successfully structural design results of RC beam.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.2
• /
• pp.151-164
• /
• 2023

Recently, the development of quantum computers means a great threat to existing public key system based on discrete algebra problems or factorization problems. Accordingly, NIST is currently in the process of contesting and screening PQC(Post Quantum Cryptography) that can be implemented in both the computing environment and the upcoming quantum computing environment. Among them, SIKE is the only Isogeny-based cipher and has the advantage of a shorter public key compared to other PQC with the same safety. However, like conventional cryptographic algorithms, all quantum-resistant ciphers must be safe for existing cryptanlysis. In this paper, we studied power analysis-based cryptographic analysis techniques for SIKE, and notably we analyzed SIKE through wavelet transformation and deep learning-based clustering power analysis. As a result, the analysis success rate was close to 100% even in SIKE with applied masking response techniques that defend the accuracy of existing clustering power analysis techniques to around 50%, and it was confirmed that was the strongest attack on SIKE.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.3
• /
• pp.112-118
• /
• 2023

As technology advances, the need for enhanced preparedness against cyber-attacks becomes an increasingly critical problem. Therefore, it is imperative to consider various circumstances and to prepare for cyber-attack strategic technology. This paper proposes a method to solve network security problems by applying reinforcement learning to cyber-security. In general, traditional static cyber-security methods have difficulty effectively responding to modern dynamic attack patterns. To address this, we implement cyber-attack scenarios such as 'Tiny Alpha' and 'Small Alpha' and evaluate the performance of various reinforcement learning methods using Network Attack Simulator, which is a cyber-attack simulation environment based on the gymnasium (formerly Open AI gym) interface. In addition, we experimented with different RL algorithms such as value-based methods (Q-Learning, Deep-Q-Network, and Double Deep-Q-Network) and policy-based methods (Actor-Critic). As a result, we observed that value-based methods with discrete action spaces consistently outperformed policy-based methods with continuous action spaces, demonstrating a performance difference ranging from a minimum of 20.9% to a maximum of 53.2%. This result shows that the scheme not only suggests opportunities for enhancing cybersecurity strategies, but also indicates potential applications in cyber-security education and system validation across a large number of domains such as military, government, and corporate sectors.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.34 no.4
• /
• pp.222-228
• /
• 2023

There are two categories of online crimes related to children and adolescents: those committed by adolescents and those committed against children and adolescents. While recent trends in criminal law show consensus on strengthening punishment in cases of crimes against children and adolescents, there are mixed stances in cases of juvenile delinquency. One perspective emphasizes strict punishment, whereas the other emphasizes dispositions aligned with human rights. While various forms of online crime share the commonality in that the main part of the criminal act occurs online, they can be categorized into three types: those seeking financial gain, those driven by sexual motives, and those engaged in bullying. Among these, crimes driven by sexual motives are the most serious. Second-hand trading fraud and conditional (sexual) meeting fraud fall under the category of seeking financial gain and occur frequently. Crimes driven by sexual motives include obscenity via telecommunication, filming with discrete cameras, child and adolescent sexual exploitation material, fake video distribution, and blackmail/coercion using intimate images/videos ("sextortion"). These crimes lead to various legal issues such as whether to view vulgar acronyms or body cams that teenagers frequently use as simple subcultures or crimes, what criteria should be applied to judge whether a recorded material induces sexual desire or shame, and at what stage sexual grooming becomes punishable. For example, sniping posts, KakaoTalk prisons, and chat room explosions are tricky issues, as they may or may not be punished depending on the case. Particular caution should be exercised against the indiscriminate application of a strict punishment-oriented approach to the juvenile justice system, which is being discussed in relation to online sexual offenses. In the punishment case of online crime, juvenile offenders with a high potential for future improvement and reform must be treated with special consideration.

• International Journal of Computer Science & Network Security
• /
• v.24 no.2
• /
• pp.158-168
• /
• 2024

There is hardly any person in modern times who has not taken soft drinks instead of drinking water. The rate of people taking soft drinks being surprisingly high, researchers around the world have cautioned from time to time that these drinks lead to weight gain, raise the risk of non-communicable diseases and so on. Therefore, in this work an image-based tool is developed to monitor the nutritional information of soft drinks by using deep convolutional neural network with transfer learning. At first, visual saliency, mean shift segmentation, thresholding and noise reduction technique, collectively known as 'pre-processing' are adopted to extract the location of drinks region. After removing backgrounds and segment out only the desired area from image, we impose Discrete Wavelength Transform (DWT) based resolution enhancement technique is applied to improve the quality of image. After that, transfer learning model is employed for the classification of drinks. Finally, nutrition value of each drink is estimated using Bag-of-Feature (BoF) based classification and Euclidean distance-based ratio calculation technique. To achieve this, a dataset is built with ten most consumed soft drinks in Bangladesh. These images were collected from imageNet dataset as well as internet and proposed method confirms that it has the ability to detect and recognize different types of drinks with an accuracy of 98.51%.

• Journal of the Korea Society for Simulation
• /
• v.22 no.4
• /
• pp.39-47
• /
• 2013

Simulation based Battle Experimentation is to examine the readiness for a battle using simulation technology. It heavily relies on the weapon systems modeling and simulation. To analyze the characteristics and complexity of the weapon systems in the experiment, the modeling & simulation environment has to be able to break down the system of systems into components and make the use of high fidelity components such as real hardware in simulation. In that sense, the modular and hierarchical structure of DEVS (Discrete EVent System Specification) framework provides potentials to meet the requirements of the battle experimentation environment. This paper describes the development of the DEVS integrated development environment for Simulation based Battle Experimentation. With the design principles of easy, flexible, and fast battle simulation, the newly developed battle experimentation tool mainly consists of 3 parts - model based graphical design tool for making DEVS models and linking them with external simulators easily through diagrams, the experiment plan tool for speeding up a statistic analysis, the standard components model libraries for lego-like building up a weapon system. This noble simulation environment is to provide a means to analyze complex simulation based experiments with different levels of models mixed in a simpler and more efficient way.

• Asian-Australasian Journal of Animal Sciences
• /
• v.22 no.3
• /
• pp.356-364
• /
• 2009

The current ruminant feeding models require parameterization of the digestion kinetics of carbohydrate fractions in feed ingredients to estimate the supply of nutrients from a ration. Using an automated gas production technique, statistically welldefined digestion rate of carbohydrate, including soluble carbohydrate, can be estimated in a relatively easy way. In this study, the gas production during in vitro fermentation was measured and recorded by an automated gas production system to investigate degradation kinetics of carbohydrate fractions of a wide range of ruminant feeds: corn silage, rice straw, corn, soybean hull, soybean meal, and cell mass from lysine production (CMLP). The gas production from un-fractionated, ethanol insoluble residue and neutral detergent insoluble residue of the feed samples were obtained. The gas profiles of carbohydrate fractions on the basis of the carbohydrate scheme of the Cornell Net Carbohydrate and Protein System (A, B1, B2, B3 and C) were generated using a subtraction approach. After the gas profiles were plotted with time, a curve was fitted with a single-pool exponential equation with a discrete lag to obtain kinetic parameters that can be used as inputs for modern nutritional models. The fractional degradation rate constants (Kd) of corn silage were 11.6, 25.7, 14.8 and 0.8%/h for un-fractioned, A, B1 and B2 fractions, respectively. The values were statistically well estimated, assessed by high t-value (>12.9). The Kd of carbohydrate fractions in rice straw were 4.8, 21.1, 5.7 and 0.5%/h for un-fractioned, A, B1 and B2 fractions, respectively. Although the Kd of B2 fraction was poorly defined with a t-value of 4.4, the Kd of the other fractions showed tvalues higher than 21.9. The un-fractioned corn showed the highest Kd (18.2%/h) among the feeds tested, and the Kd of A plus B1 fraction was 18.7%/h. Soybean hull had a Kd of 6.0, 29.0, 3.8 and 13.8%/h for un-fractioned, A, B1 and B2, respectively. The large Kd of fraction B2 indicated that NDF in soybean hull was easily degradable. The t-values were higher than 20 except for the B1 fraction (5.7). The estimated Kd of soybean meal was 9.6, 24.3, 5.0%/h for un-fractioned, A and B1 fractions, respectively. A small amount of gas (5.6 ml at 48 ho of incubation) was produced from fermentation of CMLP which contained little carbohydrate. In summary, the automated gas production system was satisfactory for the estimation of well defined (t-value >12) kinetic parameters and Kd of soluble carbohydrate fractions of various feedstuffs that supply mainly carbohydrate. The subtraction approach, however, should be applied with caution for some concentrates, especially those which contain a high level of crude protein since nitrogen-containing compounds can interfere with gas production.

• Journal of Korea Multimedia Society
• /
• v.9 no.2
• /
• pp.139-150
• /
• 2006

A schematic map is a special purpose map which is generated to recognize it's objects easily and conveniently via simplifying and highlighting logical geometric information of a map. To manufacture the schematic map with road, label and icon, we must generate simplified route map and replace many geometric objects. Performing a give task, however, there are an amount of overlap areas between geometric objects whenever we process the replacement of geometry objects. Therefore we need replacing geometric objects without overlap. But this work requires much computational resources, because of the high complexity of the original geometry map. We propose the schematic map generation system whose map consists of icons and label. The proposed system has following steps: 1) eliminating kinks that are least relevant to the shape of polygonal curve using DCE(Discrete Curve Evolution) method. 2) making an evenly distributed route using CVT(Centroidal Voronoi Tessellation) and Grid snapping method. Therefore we can keep the structural information of the route map from CVT method. 3) replacing an icon and label information with collision avoidance algorithm. As a result, we can replace the vertices with a uniform distance and guarantee the available spaces for the replacement of icons and labels. We can also minimize the overlap between icons and labels and obtain more schematized map.

• Journal of the Korean Geosynthetics Society
• /
• v.9 no.1
• /
• pp.47-57
• /
• 2010

Granular soils having a large particle size have been used as a filling material in the construction of foundation, harbor, dam, and so on. Consequently, the shear behavior of this granular soil plays a key role in respect of stability of structures. For example, soil particle crushing occurring at the interface between structure and soil and/or within soil mass can cause a disturbance of ground characteristics and consequently induce issues in respect of stability of structures. In order to investigate the shear behavior according to an existence and nonexistence of particle crushing, numerical analyses were conducted by using the DEM (Discrete Element Method)-based software program PFC2D (Particle Flow Code). By dividing soil particle bonding model into crushing model and noncrushing model, total four particle bonding models were simulated and their results were compared. Noncrushing model included one ball model and clump model, and crushing model included cluster model and Lobo-crushing model. The combinations of soil particle followed the research results of Lobo-Guerrero and Vallejo (2005) which were composed of eight circles. The results showed that the friction angle was in order of clump model > cluster model > one ball model. The particle bonding model compared to one ball model and noncrushing model compared to crushing model showed higher shear strength. It was also concluded that the model suggested by Lobo-Guerrero and Vallejo (2005) is not appropriate to simulate the soil particle crushing.

• Korean Journal of Occupational Health Nursing
• /
• v.15 no.1
• /
• pp.14-29
• /
• 2006

Purpose: The purposes of this study were to survey the extent of pain and discomfort in the musculoskeletal system among visiting housekeepers, above all concerning neck, shoulder, back, wrist, knee, and arm pain and to find possible relations between symptoms and various working conditions. Method: A questionnaire was answered by 174 woman visiting housekeepers living in Kyeonggi-do and Seoul from December 1, 2003 to February 30, 2004. The symptoms of musculoskeletal system were coded by the pain index which illustrates the extent of the symptoms, and analyzed in view of NIOSH guideline and Kim, et. al.'s notion. Result: 1. As to the complaint rate of subjective musculoskeletal symptoms by body region, the figure was the highest for shoulder with 78.2%, followed by back with 66.7%, knee 53.6%, neck 56.3%, wrist 40.2%, and arm 29.2%, respectively. The logistic analysis showed shoulder pain and arm pain have no relation with working and health conditions, and back pain was significantly related to current health condition. In same way, knee pain and wrist pain were found to be mainly related to marital status. 2. Following the NIOSH guideline, the positive rate of subjective musculoskeletal symptoms was found out in following order: shoulder 69.5%, back 59.2%, knee 54%, neck 46%, wrist 32.8%, and arm 25.3%. To investigate the main cause of each disease, the symptoms were classified by pain index, where the value of more than 3 comes to the NIOSH case, and analyzed in term of complaint rate using discrete logistical method : shoulder pain was highly related to the housekeeping time after work, back pain was to current health condition and the heavy weight carrying and neck, wrist, arm pain were commonly related to the ordinary health condition. For knee pain, working speed was a main cause. 3. In view of Kim et. al.'s standard, where the pain index is over 7, the positive rate was showed in order slightly different from previous analyses : shoulder 33.3%. knee 29.9%, back 28.2%, neck 17.2%, wrist 17.2%, and 16.7%. From the logistical analysis, insufficient rest was shown as the main cause of shoulder, back, arm and wrist pain. For neck pain, ordinary health condition was mainly related. In case of knee pain, any apparent relation is not found. Conclusion: According to the logistic regression analysis of musculoskeletal system, there was strong suggestion that the less insufficient physical rest, the more significant disorder complaint. This means that the most musculoskeletal symptom among the visiting housekeepers can be prevented and cured by sufficient physical resting.