• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.715-718
• /
• 2018

최근 다층의 인공신경망 모델이 수많은 분야에 대한 해결 방안으로 제시되고 있으며 2015년 Mnih이 고안한 DQN(Deep Q Network)는 Atari game에서 인간 수준의 성능을 보여주며 많은 이들에게 놀라움을 자아냈다. 본 논문에서는 Atari DQN Model을 실내 자율주행 모바일 로봇에 적용하여 신경망 모델이 최단 경로를 추종하며 장애물 회피를 위한 행동을 학습시키기 위해 로봇이 가지는 상태 정보들을 84*84 Mat로 가공하였고 15가지의 행동을 정의하였다. 또한 Virtual world에서 신경망 모델이 실제와 유사한 현재 상태를 입력받아 가장 최적의 정책을 학습하고 Real World에 적용하는 방법을 연구하였다.

• The Journal of the Korea Contents Association
• /
• v.17 no.6
• /
• pp.203-224
• /
• 2017

With the development of the IT industry and the growth of the cultural industry, the game industry is becoming an important industry. In this regard, the study seeks to differentiate the generation of video games based on technological characteristics from the perspective of technological innovation. SEGA, Nintendo, MicroSoft, SONY, and ATARI were chosen as research subjects. The survy was conducted from ATARI to 2017. The results of the study are expected to help develop the technology strategy of the future video game industry.

• Journal of Korean Electronics
• /
• v.4 no.3
• /
• pp.61-67
• /
• 1984

• Proceedings of the Korea Contents Association Conference
• /
• 2005.05a
• /
• pp.64-69
• /
• 2005

Recently, grafting of a story into a game has attracted ever increasing attention. In this study, the 12 phases of Hero's Journey by Joseph Campbell highlighted in movies and myths was analysed from a point of Christoper Vogler to be correlated with a game. Since a story-teller Is a player in a game, the story experienced by the player enjoying the game was examined and correlated with the 12 phases of Campbell. As the game, a construction/management simulation game, RCT3 produced by Atari was studied. The 12 phases of Campbell was applicable to game story.

• The Journal of the Korea Contents Association
• /
• v.6 no.11
• /
• pp.104-109
• /
• 2006

Recently, embedding method of a story into a game has gathered much attention than ever. In this study, the 12 phases of Hero's Journey, which was suggested by Joseph Campbell and was highlighted in the area of movies and myths, has been analysed at a point of Christoper Vogler. Since a gamer is not just only a player but also a story-teller in a game, the story experienced by the player enjoying the game was examined and correlated with the 12 phases of Campbell. The game RCT3, a construction/management simulation game ,produced by Atari, was studied. The 12 phases of Campbell was applicable to game story.

• Convergence Security Journal
• /
• v.15 no.7
• /
• pp.63-68
• /
• 2015

In rapid aging society, the disease prevention and management for healthy life of the elderly is very important. Regular physical activity is known to be a useful intervention for the physical and mental health of the elderly. In this study, we explore the convergence contents using IT technology as the intervention for encouraging the regular physical activity in the elderly. There are u-Healthcare and serious game as the convergence of health care and IT technology, and the serious game which is added special purpose such as education, training, and treatment to the fun-one of the game element- can be suitable to provide a variety of contents that leads to physical activity in the elderly. The contents inducing physical activities are "Puffer(ATARI, USA)", "WiiFit(Nintendo, JAPAN)", "Age Invaders(MXR Lab, SINGAPORE)", "Xbox $360^{\circ}$+kinect(Microsoft, USA)", "Tangible bicycle game(Donsin Univ., KOREA)", and "3D Gateball game(Soongsil Univ., KOREA)", and these contents can help health care of the elderly. By increasing physical activity through the use of these contents, it will be able to promote physical fitness and body function required in daily life, disease prevention, and maintain health in the elderly.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.5
• /
• pp.155-162
• /
• 2021

n-step TD learning is a combination of Monte Carlo method and one-step TD learning. If appropriate n is selected, n-step TD learning is known as an algorithm that performs better than Monte Carlo method and 1-step TD learning, but it is difficult to select the best values of n. In order to solve the difficulty of selecting the values of n in n-step TD learning, in this paper, using the characteristic that overestimation of Q can improve the performance of initial learning and that all n-step returns have similar values for Q ≈ Q^*, we propose a new learning target, which is composed of the maximum and the mean of all k-step returns for 1 ≤ k ≤ n. Finally, in OpenAI Gym's Atari game environment, we compare the proposed algorithm with n-step TD learning and proved that the proposed algorithm is superior to n-step TD learning algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.853-873
• /
• 2021

In recent years, deep reinforcement learning (DRL) models are enjoying great interest as their success in a variety of challenging tasks. Deep Q-Network (DQN) is a widely used deep reinforcement learning model, which trains an intelligent agent that executes optimal actions while interacting with an environment. This model is well known for its ability to surpass skilled human players across many Atari 2600 games. Although DQN has achieved excellent performance in practice, there lacks a clear understanding of why the model works. In this paper, we present a visual analytics system for understanding deep Q-network in a non-blind matter. Based on the stored data generated from the training and testing process, four coordinated views are designed to expose the internal execution mechanism of DQN from different perspectives. We report the system performance and demonstrate its effectiveness through two case studies. By using our system, users can learn the relationship between states and Q-values, the function of convolutional layers, the strategies learned by DQN and the rationality of decisions made by the agent.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.141-148
• /
• 2022

Recently, reinforcement learning combined with deep neural network technology has achieved remarkable success in various fields such as board games such as Go and chess, computer games such as Atari and StartCraft, and robot object manipulation tasks. However, such deep reinforcement learning describes states, actions, and policies in vector representation. Therefore, the existing deep reinforcement learning has some limitations in generality and interpretability of the learned policy, and it is difficult to effectively incorporate domain knowledge into policy learning. On the other hand, dNL-RRL, a new relational reinforcement learning framework proposed to solve these problems, uses a kind of vector representation for sensor input data and lower-level motion control as in the existing deep reinforcement learning. However, for states, actions, and learned policies, It uses a relational representation with logic predicates and rules. In this paper, we present dNL-RRL-based policy learning for transportation mobile robots in a manufacturing environment. In particular, this study proposes a effective method to utilize the prior domain knowledge of human experts to improve the efficiency of relational reinforcement learning. Through various experiments, we demonstrate the performance improvement of the relational reinforcement learning by using domain knowledge as proposed in this paper.

• Archives of Plastic Surgery
• /
• v.43 no.4
• /
• pp.365-370
• /
• 2016

Background There is a growing concern in the field of aesthetic surgery about the need to measure patients' expectations preoperatively. The present study was designed to develop and validate the Expectations of Aesthetic Rhinoplasty Scale (EARS), and to compare expectations between rhinoplasty patients with and without body dysmorphic disorder (BDD). Methods In total, 162 college students and 20 rhinoplasty candidates were recruited. The measures included the newly developed EARS, a measure of psychopathology, and demographics. The DSM-IV structured clinical interview for BDD was used to confirm the diagnosis in rhinoplasty patients. Results The EARS was constructed of six items based on their significant content validity. In the scale development phase, Cronbach's alpha was 0.87. The test-retest reliability coefficient of the scale was satisfactory (intraclass correlation coefficient, 0.94; 95% confidence interval, 0.82-0.98) over a four-week period. Scores on the EARS were significantly positively correlated with psychopathological symptoms (r=0.16; P<0.05). Moreover, comparison of EARS scores between BDD (M=25.90, standard deviation [SD]=6.91) and non-BDD rhinoplastic patients (M=15.70, SD=5.27) suggested that BDD patients held significantly higher expectations (P<0.01). Conclusions The expectations of aesthetic rhinoplasty patients toward surgery may play a crucial role in their postoperative satisfaction. While the value of patients' expectations is clinically recognized, no empirical study has measured these expectations in a psychometrically sound manner. The current study developed and validated the EARS. It may be easily used as a valid and reliable instrument in clinical and research settings.