• Title/Summary/Keyword: CAA

Search Result 147, Processing Time 0.028 seconds

L-CAA : An Architecture for Behavior-Based Reinforcement Learning (L-CAA : 행위 기반 강화학습 에이전트 구조)

  • Hwang, Jong-Geun;Kim, In-Cheol
    • Journal of Intelligence and Information Systems
    • /
    • v.14 no.3
    • /
    • pp.59-76
    • /
    • 2008
  • In this paper, we propose an agent architecture called L-CAA that is quite effective in real-time dynamic environments. L-CAA is an extension of CAA, the behavior-based agent architecture which was also developed by our research group. In order to improve adaptability to the changing environment, it is extended by adding reinforcement learning capability. To obtain stable performance, however, behavior selection and execution in the L-CAA architecture do not entirely rely on learning. In L-CAA, learning is utilized merely as a complimentary means for behavior selection and execution. Behavior selection mechanism in this architecture consists of two phases. In the first phase, the behaviors are extracted from the behavior library by checking the user-defined applicable conditions and utility of each behavior. If multiple behaviors are extracted in the first phase, the single behavior is selected to execute in the help of reinforcement learning in the second phase. That is, the behavior with the highest expected reward is selected by comparing Q values of individual behaviors updated through reinforcement learning. L-CAA can monitor the maintainable conditions of the executing behavior and stop immediately the behavior when some of the conditions fail due to dynamic change of the environment. Additionally, L-CAA can suspend and then resume the current behavior whenever it encounters a higher utility behavior. In order to analyze effectiveness of the L-CAA architecture, we implement an L-CAA-enabled agent autonomously playing in an Unreal Tournament game that is a well-known dynamic virtual environment, and then conduct several experiments using it.

  • PDF

Pathogenicity of a Local Isolate of Chicken Anemia Agent for Chickens and Prevalence of Antibody in Chicken Flocks (국내분리 chicken anemia agent의 닭에 대한 병원성과 야외계군의 항체 보유상황)

  • 김선중
    • Korean Journal of Poultry Science
    • /
    • v.18 no.3
    • /
    • pp.141-150
    • /
    • 1991
  • A local isolate of chicken anemia agent (CAA), isolate 89-69. was tested for pathogenicity for chickens. When chickens from a specific pathogen free (SPF) flock were inoculated intramuscularly with the isolate at one day old, all the chickens showed severe anemia at 14 to 18 days post inoculation(DPI) and returned to normal at 25DPI, Some of the inoculated chickens (27∼33%) died between 13 to 17 DPI's with lesions of severe aplasia of bone marrow and thymic atrophy. In chickens kept in contact with inoculated chickens, some of the chickens had anemia at 25 and 28 DPI's. Virus could be reisolated from inoculated as well as in contact chickens till 21 DPI. Antibodies to CAA could be detected in all inoculated and in contact chickens when tested at 42 DPI by the indirect fluorescent antibody method. When chickens from a different SPF flock were inoculated at one day old, degrees of anemia, both in frequency of incidence and severity, were low These chickens were proved partly to have antibodies to CAA when tested for hatchmates. In a survey for antibodies to CAA in field chicken flocks, one out of 7 flocks(14%) aged 3 to 10weeks was antibody positive whereas 19 out of 20 flocks(95%) over 20 weeks of age were positive. Altogether 29 out of 39 flocks (74%) were antibody positive.

  • PDF

An Agent Architecture for Behavior-Based Reinforcement Learning (행위 기반 강화 학습 에이전트 구조)

  • Hwang, Jong-Geun;Kim, In-Cheol
    • Proceedings of the Korea Inteligent Information System Society Conference
    • /
    • 2007.11a
    • /
    • pp.284-293
    • /
    • 2007
  • 본 논문에서는 실시간 동정 환경에 효과적인 L-CAA 에이전트 구조를 제안한다. L-CAA 에이전트 구조는 변화하는 환경에 대한 적응성을 높이기 위해, 선행 연구를 통해 개발된 행위 기반 에이전트 구조인 CAA에 강화 학습 기능을 추가하여 확장한 것이다. 안정적인 성능을 위해 L-CAA에서 행위 선택 메커니즘은 크게 두 단계로 나뉜다. 첫 번째 단계에서는 사용자가 미리 정의한 각 행위의 수행 가능 조건과 효용성을 검사함으로써 행위 라이브러리로부터 실행할 행위들을 추출한다. 하지만 첫 번째 단계에서 다수의 행위가 추출되면, 두 번째 단계에서는 강화 학습의 도움을 받아 이들 중에서 실행할 하나의 행위를 선택한다. 즉, 강화 학습을 통해 갱신된 각 행위들의 Q 함수 값을 서로 비교함으로써, 가장 큰 기대 보상 값을 가진 행위를 선택하여 실행한다. 또한 L-CAA에서는 실행 중인 행위의 유지 가능 조건을 지속적으로 검사하여 환경의 동적 변화로 인해 일부 조건이 만족되지 않는 경우가 발생하면 현재 행위의 실행을 즉시 종료할 수 있다. 그 뿐 아니라, L-CAA는 행위 실행 중에도 효용성이 더 높은 다른 행위가 발생하면 현재의 행위를 일시 정지하였다가 복귀하는 기능도 제공한다. 본 논문에서는 L-CAA 구조의 효과를 분석하기 위해, 대표적인 동적 가상환경인 Unreal Tournament 게임에서 자율적을 동작하는 L-CAA기반의 UTBot 들을 구현하고, 이들을 이용하여 성능실험을 전개해본다.

  • PDF

Iron deficiency anemia as a predictor of coronary artery abnormalities in Kawasaki disease

  • Kim, Sohyun;Eun, Lucy Youngmin
    • Clinical and Experimental Pediatrics
    • /
    • v.62 no.8
    • /
    • pp.301-306
    • /
    • 2019
  • Purpose: Coronary artery abnormalities (CAA) are the most important complications of Kawasaki disease (KD). Iron deficiency anemia (IDA) is a prevalent micronutrient deficiency and its association with KD remains unknown. We hypothesized that presence of IDA could be a predictor of CAA. Methods: This retrospective study included 173 KD patients, divided into 2 groups according to absence (group 1) and presence (group 2) of CAA. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using a logistic regression model to estimate the association between CAA and other indicators. Due to collinearity between indicators of IDA, each indicator was paired with anemia in 3 models. Results: Serum iron, iron saturation, and ferritin concentration, the 3 indicators of IDA, were significantly higher in group 1 than in group 2. Three sets of models including anemia with iron indicators produced the OR of CAA of 3.513, 3.171, and 2.256, respectively. The 3 indicators of IDA were negatively associated with CAA, by OR of 0.965, 0.914, and 0.944, respectively. The areas under the curve (AUCs) of ferritin concentration, iron saturation, serum iron, anemia, and Kobayashi score were 0.907 (95% CI, 0.851-0.963), 0.729 (95% CI, 0.648-0.810), 0.711 (95% CI, 0.629-0.793), 0.638 (95% CI, 0.545-0.731), and 0.563 (95% CI, 0.489-0.636), respectively. Conclusion: Indicators of IDA, especially ferritin, were highly associated with CAA; therefore, they were stronger predictors of CAA than Kobayashi scores. IDA indicators can be used to predict CAA development and to suggest requirements for early interventions.

Flow and Flow Noise Analysis of HSM by Using CAA++ (CAA++를 이용한 HSM에 대한 유동과 유동소음 해석)

  • Kim, Young Nam;Chae, Jun Hee
    • Transactions of the Korean Society for Noise and Vibration Engineering
    • /
    • v.24 no.3
    • /
    • pp.205-212
    • /
    • 2014
  • In this paper, sunroof buffeting analysis for Hyundai simple model(HSM) is studied computationally. For validation, the velocity profile of boundary layer around the opening of HSM was obtained and compared with experimental results. The analysis of sunroof buffeting is done in two parts. First a steady state solution is obtained using the Reynolds Averaged Navier Stokes (RANS) solver, and then the computed flow field information is used as input for CAA++. Second transient simulation by CAA++ is performed for the peak sound pressure levels and peak frequencies of buffeting noise over the ranges of flow velocities. The benchmark results of frequency and sound pressure levels showed the general phenomena and matched well with the experimental data obtained by Hyundai Motor Car.

A Study on the Cloud Service Model of CaaS Based on the Object Identification, ePosition, with a Structured Form of Texts (문자열로 구조화된 사물식별아이디 이포지션(ePosition) 기반의 클라우드 CaaS(Contents as a Service) 서비스 모델에 관한 연구)

  • Lee, Sang-Zee;Kang, Myung-Su;Cho, Won-Hee
    • Information Systems Review
    • /
    • v.15 no.3
    • /
    • pp.129-139
    • /
    • 2013
  • The Internet of Things (or IoT for short) which refers to uniquely identifiable objects and their virtual representations in an Internet-like structure is to be reality today. The amount of data on IoT is expected to increase abruptly and there are several key issues like usefulness interoperability between multiple distributes systems, services and databases. In this paper a methodology is proposed to realize a recently developed cloud service model, Contents as a Service (CaaS), which is contents delivery model referred to as 'on-demand contents'. In the proposed method, the global object identification, ePosition, comprising the structured form of two sorts of text strings with a separation symbol like # is applied to identify a specific content and registered with the content at the same server. It is easy-to-realize and effective to solve the interoperability problem systematically and logically. Some APIs for the proposed CaaS service are to be converged to provide some upgraded cloud service model such as 'CaaS supported SaaS' and 'CaaS supported PaaS'.

  • PDF

Introduction to the Computational AeroAcoustics and Its Applications (전산공력음향학(CAA) 소개 및 응용사례)

  • Lee Duck-Joo;Shim In-Bo;Heo Dae-Nyoung
    • 한국전산유체공학회:학술대회논문집
    • /
    • 2000.05a
    • /
    • pp.1-13
    • /
    • 2000
  • This paper presents a review of the methodology, problems and progress in computational aeroacoustics(CAA). The nature, characteristics, and objectives of aeroacoustics problems are quite different from the commonly encountered CFD problems. In this paper, computational methods that are designed especially for CAA applications are introduced. The potential offered by CAA, the numerical issues which need to be given careful attention, and some of the recent progress in solving aeroacoustic problem are discussed

  • PDF

The Effect of the Curcumae Longae Rhizoma (CLR) Extract on the Hepatocellular Carcinogenesis and Acute Liver Damage Induced by Diethylnitrosamine (DENA) and CCl4 in Rats (강황추출물이 Diethylnitrosamine과 CCl4로 유발된 흰쥐의 간암과 간 손상에 미치는 영향)

  • Jung, Tae-San;Choi, Chang-Won
    • Herbal Formula Science
    • /
    • v.22 no.1
    • /
    • pp.177-192
    • /
    • 2014
  • Objective : In order to investigate the effect of Curcumae Longae Rhizoma(CLR) extract on the hepatocellular carcinogenesis and acute liver damage induced by diethylnitrosamine(DENA) and $CCl_4$ in rats. Methods : Experimental groups were subdivided into four; normal group (Nor), acute liver damage and hepatocellular cancer inducing control group (Con), and CLR extract 200mg/kg/day (CAA) or 400mg/kg/day (CAB) administered groups to Con. Thereafter the changes of the body weight, the liver weight and the weight of liver/100g body weight, total cholesterol, HDL cholesterol, triglyceride, the activities of AST, ALT, ALP, LDH, AFP, SOD, catalase were measured. And we observed by optical and electron microscopy. Result : 1. The body weight was decreased in Con compared with Nor for 5 weeks, but increased in Con compared with Nor from 6 week to 9 week. During experimental period of total 9 weeks, CAA and CAB were increased compared with Con. 2. The liver weight was increased significantly (p<0.05) in Con compared with Nor. The weight of liver/100g body weight was increased significantly (p<0.05) in Con compared with Nor and decreased significantly (p<0.05) in CAB compared with Con. 3. The level of total cholesterol was increased in Con and CAA compared with Nor, but there was not statistically significant. The level of triglyceride was decreased in Con compared with Nor. But increased in CAA and CAB compared with Con. The level of HDL-cholesterol was significantly increased (p<0.05) in CAA and CAB compared with Con. 4. The activities of AST, ALT were increased in Con compared with Nor, but decreased in CAA compared with Con, significantly decreased (p<0.05) in CAB compared with Con. 5. The activities of ALP, LDH were increased in Con compared with Nor, but decreased in CAA and CAB compared with Con. 6. The activities of AFP was increased significantly (p<0.05) in Con compared with Nor, but decreased significantly (p<0.05) in CAA and CAB compared with Con. 7. The activities of SOD were increased in Con, CAA and CAB compared with Nor, but decreased in CAA and CAB compared with Con. The activities of Catalase was more increased in CAA and CAB compared than Con. 8. The results of light microscopical observation, a number of hepatocytes were damaged in Con compared with Nor and CAB. 9. According to the electron microscopical observation, irregular nuclear membrane, condensed nucleoplasm was observed in Con, the experimental group was observed in the nucleus of the well-preserved and evenly developed nucleoplasm. Conclusions : These results suggest that administration of CLR extract suppress or retard on the hepatocellular carcinogenesis and acute liver damage induced by DENA and $CCl_4$ in rats.