• Journal of Semiconductor Engineering
• /
• v.2 no.2
• /
• pp.130-135
• /
• 2021

Recently, deep neural networks (DNNs) are actively used for action control so that an autonomous system, such as the robot, can perform human-like behaviors and operations. Unlike recognition tasks, the real-time operation is essential in action control, and it is too slow to use remote learning on a server communicating through a network. New learning techniques, such as reinforcement learning (RL), are needed to determine and select the correct robot behavior locally. In this paper, we propose an energy-efficient DNN processor with a LUT-based processing engine and near-zero skipper. A CNN-based facial emotion recognition and an RNN-based emotional dialogue generation model is integrated for natural HRI system and tested with the proposed processor. It supports 1b to 16b variable weight bit precision with and 57.6% and 28.5% lower energy consumption than conventional MAC arithmetic units for 1b and 16b weight precision. Also, the near-zero skipper reduces 36% of MAC operation and consumes 28% lower energy consumption for facial emotion recognition tasks. Implemented in 65nm CMOS process, the proposed processor occupies 1784×1784 um2 areas and dissipates 0.28 mW and 34.4 mW at 1fps and 30fps facial emotion recognition tasks.

• The Journal of the Korea Contents Association
• /
• v.14 no.7
• /
• pp.18-28
• /
• 2014

With the reformation of organization and function of local office of education in 2010, consulting-supervision is introduced to schools as a system for education reform to improve the quality of school education. However, a dedicated on-line portal system that can provide integrated management on the functionalities of consulting-supervision has not been implemented yet. To successfully operate consulting-supervision in schools, it is also needed to provide an on-line consultant education system, that can support teachers to train themselves as a supervision-consultant. In this paper, we introduce an on-line consultant training system that provides various learning activity tools for consultant training based on Learning Activity Management System(LAMS) and Action Learning. The system consists of Management stage, Analysis stage, Solution stage, and Action stage for the empowerment of consultants' expertises, and is named as MASA. Brain-writing, SWOT(Strengths, Weaknesses, Opportunities, and Threads) analysis, 5Whys, decision grid, PMI(Plus, Minus, Interesting), and black chart techniques were developed in MASA as learning activity tools for consultant training.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2614-2616
• /
• 2002

According to the appearance of various virtual websites using multimedia technologies for engineering education, the internet applications in engineering education have drawn much interests. But unidirectional communication, simple text/image based webpages and tedious learning process without motivation etc. have made the lowering of educational efficiency in cyberspace. Thus, to cope with these difficulties this paper presents a web-based educational Flash movies based on ActionScript language for understanding the principles of the computer system architecture. The proposed Flash movies provides the improved learning methods which can enhance the interests of learners. The results of this paper can be widely used to improve the efficiency of cyberlectures in the cyber university. Several sample Flash movies are illustrated to show the validity of the proposed learning method.

• The Journal of Korea Robotics Society
• /
• v.17 no.1
• /
• pp.40-47
• /
• 2022

A reward function suitable for a task is required to manipulate objects through reinforcement learning. However, it is difficult to design the reward function if the ample information of the objects cannot be obtained. In this study, a demonstration-based object manipulation algorithm called stochastic exploration guided by demonstration (SEGD) is proposed to solve the design problem of the reward function. SEGD is a reinforcement learning algorithm in which a sparse reward explorer (SRE) and an interpolated policy using demonstration (IPD) are added to soft actor-critic (SAC). SRE ensures the training of the critic of SAC by collecting prior data and IPD limits the exploration space by making SEGD's action similar to the expert's action. Through these two algorithms, the SEGD can learn only with the sparse reward of the task without designing the reward function. In order to verify the SEGD, experiments were conducted for three tasks. SEGD showed its effectiveness by showing success rates of more than 96.5% in these experiments.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.426-432
• /
• 2022

HVAC systems play a critical role in reducing energy consumption in buildings. Integrating occupants' thermal comfort evaluation into HVAC control strategies is believed to reduce building energy consumption while minimizing their thermal discomfort. Advanced technologies, such as visual sensors and deep learning, enable the recognition of occupants' discomfort-related actions, thus making it possible to estimate their thermal discomfort. Unfortunately, it remains unclear how accurate a deep learning-based classifier is to recognize occupants' discomfort-related actions in a working environment. Therefore, this research evaluates the classification performance of occupants' discomfort-related actions while sitting at a computer desk. To achieve this objective, this study collected RGB video data on nine college students' cold discomfort-related actions and then trained a deep learning-based classifier using the collected data. The classification results are threefold. First, the trained classifier has an average accuracy of 93.9% for classifying six cold discomfort-related actions. Second, each discomfort-related action is recognized with more than 85% accuracy. Third, classification errors are mostly observed among similar discomfort-related actions. These results indicate that using human action data will enable facility managers to estimate occupants' thermal discomfort and, in turn, adjust the operational settings of HVAC systems to improve the energy efficiency of buildings in conjunction with their thermal comfort levels.

• Journal of Advanced Navigation Technology
• /
• v.24 no.2
• /
• pp.155-162
• /
• 2020

Recently, with the development of computer vision and deep learning technology, research on human action recognition has been actively conducted for video analysis, video surveillance, interactive multimedia, and human machine interaction applications. Diverse techniques have been introduced for human action understanding and classification by many researchers using RGB image, depth image, skeleton and inertial data. However, skeleton-based action discrimination is still a challenging research topic for human machine-interaction. In this paper, we propose an end-to-end skeleton joints mapping of action for generating spatio-temporal image so-called dynamic image. Then, an efficient deep convolution neural network is devised to perform the classification among the action classes. We use publicly accessible UTD-MHAD skeleton dataset for evaluating the performance of the proposed method. As a result of the experiment, the proposed system shows better performance than the existing methods with high accuracy of 97.45%.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.8
• /
• pp.247-254
• /
• 2021

The purpose of this study is to investigate the effects of action learning on the satisfaction of majors by cultivating task-solving ability through self-reflection in the form of team learning for a certain period of time in the form of team learning by using action learning for students taking convergence curriculum in universities. The subjects of the study were 40 students from the Department of Sports Rehabilitation, a convergence of the Department of Sports and Health Management and the Department of Physical Therapy located in Jeollanam-do. This was conducted to confirm the difference in the effect of satisfaction. Comparison of changes in groups of experimental groups with action learning teaching methods showed significant differences in self-directed learning skills, problem-solving skills, and major satisfaction(p<.001)(p<.05). A significant difference in self-directed learning ability, problem-solving ability, and major satisfaction was also shown in the comparison of changes in control groups that applied traditional teaching methods(p<.05). Comparison of changes between groups showed significant differences in self-directed learning skills, problem-solving skills and major satisfaction(p<.05). Applying the action learning teaching method to the level of students in the convergence course will improve self-directed learning skills, problem-solving skills, and major satisfaction, and further research will be needed to expand the target and add variables to combine qualitative research.

• Journal of the Korea Convergence Society
• /
• v.13 no.5
• /
• pp.489-499
• /
• 2022

This study examined the effect of convergence action learning based virtual simulation program for nursing college students. The study was carried out according a nonequivalent control group design. The study subjects were 54 nursing college students. The data collection period was from April 12, 2021 to June 18, 2021. Collected data were analyzed using SPSS PC+ 23.0. The experimental group had significantly different to communication skills, and clinical competency in comparison to the control group. This suggests that the convergence action learning based virtual simulation program can be applied as a way to increase nursing students' communication skills, and clinical competency.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.31 no.3
• /
• pp.27-40
• /
• 2006

Today's customer demands in supply chains tend to change quickly, variously even in a short time Interval. The uncertainties of customer demands make it difficult for supply chains to achieve efficient inventory replenishment, resulting in loosing sales opportunity or keeping excessive chain wide inventories. Un this paper, we propose an adaptive vendor managed inventory (VMI) model for a two-echelon supply chain with non-stationary customer demands using the action-reward learning method. The Purpose of this model is to decrease the inventory cost adaptively. The control Parameter, a compensation factor, is designed to adaptively change as customer demand pattern changes. A simulation-based experiment was performed to compare the performance of the adaptive VMI model.

• Educational Technology International
• /
• v.7 no.1
• /
• pp.81-98
• /
• 2006

The purpose of this study was to apply the ADDIE instructional design model to develop multimedia rich project-based learning methods for effective instruction in a Korean mechanical engineering high school. This study was conducted as action research based on a high school situation. The study included 40 participants in a class purposively selected from 52 classes at 2080 student high school. Data were collected through observations, surveys and artifacts. Results indicated the multimedia rich project-based learning allowed students to take part in learning activities and there was close cooperation with and among group members to create better products. Also, the flexibility in the project-based learning environment allowed the participants to make decisions about their abilities, resources, and plans. Recommendations and implications for teacher educators as well as in-service and pre-service teachers are also presented.