• ETRI Journal
• /
• v.41 no.1
• /
• pp.109-116
• /
• 2019

This paper explores new techniques that are based on a hidden-layer linear transformation for fast speaker adaptation used in deep neural networks (DNNs). Conventional methods using affine transformations are ineffective because they require a relatively large number of parameters to perform. Meanwhile, methods that employ singular-value decomposition (SVD) are utilized because they are effective at reducing adaptive parameters. However, a matrix decomposition is computationally expensive when using online services. We propose the use of an extended diagonal linear transformation method to minimize adaptation parameters without SVD to increase the performance level for tasks that require smaller degrees of adaptation. In Korean large vocabulary continuous speech recognition (LVCSR) tasks, the proposed method shows significant improvements with error-reduction rates of 8.4% and 17.1% in five and 50 conversational sentence adaptations, respectively. Compared with the adaptation methods using SVD, there is an increased recognition performance with fewer parameters.

• Korean Journal of Social Welfare
• /
• v.41
• /
• pp.173-200
• /
• 2000

The purpose of this study is to examine the variables that may influence the level of family adaptation to schizophrenic patients using the Family Resiliency Model. The Family Resiliency Model is the most current extension of family stress Model. According to the Family Resiliency Model, The level of family adaptation in the face of a crisis situation is determined by a number of interacting components. The subjects are 151 family members with schizophrenic patient. The result from the research were as follows: 1) The following variables significantly correlated with the family adaptation: income of the family, educational level of the family, intimacy between family and patient, knowledge on schizophrenia, recognition of prognosis on schizophrenia. 2) The factors that compose the Family Resiliency Model significantly correlated with the level of family adaptation. 3) The result of stepwise multiple regression analysis indicated that factors which predict the level of family adaptation were the family control, the quality of family communication, and the support from the extended family, these findings give us significant practical implications for social work intervention.

• Journal of the Korea Society of Computer and Information
• /
• v.9 no.3
• /
• pp.149-155
• /
• 2004

This paper describes the self-adaptive Content Service Network (CSN) on Application Level Active Network (ALAN). Web caching technology comprises Content Delivery Network (CDN) for content distribution as well as Content Service Network (CSN) for service distribution. The IETF working group on Open Pluggalble Edge Service (OPES) is the works closely related to CSN. But it can be expected that the self-adaptation in ubiquitous computing environment will be deployed. The existing content service on CSN lacks in considering self-adaptation. This results in inability of existing network to support the additional services. Therefore, in order to address the limitations of the existing networks, this paper suggests Self-adaptive Content Service Network (CSN) using the GME and the extended ALAN to insert intelligence into the existing network.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.34 no.6
• /
• pp.969-981
• /
• 2010

This study explains the consumer acceptance of smart clothing using the extended Technology Acceptance Model (TAM); in addition, it compares the difference in the path hypotheses of the perceived group and nonperceived group from the aspect of the extended TAM. A total of 815 copies of questionnaire were collected from a web-based survey in March 2009. Structural equation modeling was used to examine the entire pattern of intercorrelations among the constructs and to test related propositions using an AMOS 5.0 package. The fitness of the extended TAM explains the process of the adaptation of smart clothing. Technology Innovation (TI) and Clothing Involvement (CI) were confirmed as antecedent variables to affect TAM. In the perceived group, Technology Innovation (TI) and Clothing Involvement (CI) showed significant impacts on the Perceived Ease of Use (PEOU) and Perceived Usefulness (PU) while Technology Innovation (TI) did not influence the Perceived Ease of Use (PEOU) in the non-perceived group. Perceived Ease of Use (PEOU) influenced the Perceived Usefulness (PU) and indirectly influenced Attitude (A) through the Perceived Usefulness (PU) in both groups. In addition, Perceived Usefulness (PU) did not influence Acceptance Intention (AI) but indirectly affected Acceptance Intention (AI) through Attitude (A). Therefore, Attitude (A) was found to be an important parameter in the adaptation of smart clothing in both groups. This finding implies that consumers first perceive the usefulness of smart clothing, then take favorable attitudes towards the smart clothing, and finally have the intention to adopt it. Strategies for publishing and informing consumers of the functions of smart clothing and usefulness in life are necessary; in addition, understanding what useful values they expect from the clothing is also crucial.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.2
• /
• pp.73-81
• /
• 2024

Recently, mobile manipulators have been utilized in agriculture industry for weed removal and harvest automation. This paper proposes a domain adaptive fruit detection method for harvest automation, by utilizing OWL-ViT model which is an open-vocabulary object detection model. The vision-language model can detect objects based on text prompt, and therefore, it can be extended to detect objects of undefined categories. In the development of deep learning models for real-world problems, constructing a large-scale labeled dataset is a time-consuming task and heavily relies on human effort. To reduce the labor-intensive workload, we utilized a large-scale public dataset as a source domain data and employed a domain adaptation method. Adversarial learning was conducted between a domain discriminator and feature extractor to reduce the gap between the distribution of feature vectors from the source domain and our target domain data. We collected a target domain dataset in a real-like environment and conducted experiments to demonstrate the effectiveness of the proposed method. In experiments, the domain adaptation method improved the AP50 metric from 38.88% to 78.59% for detecting objects within the range of 2m, and we achieved 81.7% of manipulation success rate.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.3 no.2
• /
• pp.233-238
• /
• 2003

Living creatures maintain their variety through speciation, which helps them to have more fitness for an environment. So evolutionary algorithm based on biological evolution must maintain variety in order to adapt to its environment. In this paper, we utilize the concept of speciation. Each individual of population creates their offsprings using mutation, and next generation consists of them. Each individual explores search space determined by mutation. Useful search space is extended by differentiation, then population explorers whole search space very effectively. If evolvable hardware evolves through mutation, it is useful way to explorer search space because of less varying inner structure. We verify the effectiveness of the proposed method by applying it to two optimization problems.

• Journal of Families and Better Life
• /
• v.17 no.2
• /
• pp.75-92
• /
• 1999

The purpose of this study was to investigate the caregiving and adaptation in families who awarded on filial piety. Using the qualitative methods-in depth interview genogram ecomap participation observation- 8 families were analyzed The major findings can be summarized as follows (1) These families showed very high family solidarity with strong collectivity. Their family role was interchangable and their familiarity was passed down from generation to generation. (2) Family members showed similar coping patterns on the basis of their own adaptationl Their coping was inclined toward acceptance especially religioous rather than emotional-focused ad hardly avoidance coping. (4) Although their economic emotional and service cares were family-centered with helps from their extended family household equipment for elder care was not prepared. (5) These families preferred to maintain this condition rather than to be well adapted. It was concluded that to help failed elder's family care iving social support program should be complemented in family life enrichment program for their children medical care program equipment rental program for elder in social network religion program and so on, Social norms forparent-children relation has be changed to more flexible care pattern. Also more social-emotional support has to be given to these families.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.11S
• /
• pp.3269-3277
• /
• 1999

To promote the inclusion of end-to-end congestion control in the design of future protocols using best-effort traffic, we propose a router mechanism, Extended FRED(ex-FRED). In this paper, we catagorize the TCP controlled traffics into robust and fragile traffic and discuss several unfairness conditions between them caused by the diverse applications. For example, fragile traffic from bursty application cannot use its fair share due to their slow adaptation. Ex-FRED modifies the FRED(Fair Random Early Drop), which can show wrong information due to the narrow view of actual buffer. Therefore, Ex-FRED uses per-flow accounting in larger virtual buffer to impose an each flow a loss rate that depends on the virtual buffer use of a flow. The simulation results show that Ex-FRED uses fair share and has good throughput.

• Journal of Communications and Networks
• /
• v.12 no.4
• /
• pp.346-357
• /
• 2010

Opportunistic scheduling provides the capability of resource management in wireless networks by taking advantage of multiuser diversity and by allowing delay variation in delivering data packets. It generally aims to maximize system throughput or guarantee fairness and quality of service (QoS) requirements. In this paper, we develop an extended proportional fair (PF) scheduling policy that can statistically guarantee three kinds of QoS. The scheduling policy is derived by solving the optimization problems in an ideal system according to QoS constraints. We prove that the practical version of the scheduling policy is optimal in opportunistic scheduling systems. As each scheduling policy has some parameters, we also consider practical parameter adaptation algorithms that require low implementation complexity and show their convergences mathematically. Through simulations, we confirm that our proposed schedulers show good fairness performance in addition to guaranteeing each user's QoS requirements.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.9
• /
• pp.669-678
• /
• 2003

A road-adaptive LQG control for the semi-active Macpherson strut suspension system of hydraulic type is investigated. A new control-oriented model, which incorporates the rotational motion of the unsprung mass, is used for control system design. First, based on the extended least squares estimation algorithm, a LQG controller adapting to the estimated road characteristics is designed. With computer simulations, the performance of the proposed LQC-controlled semi-active suspension is compared with that of a non-adaptive one. The results show better control performance of the proposed system over the compared one.