• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.441-452
• /
• 2021

Federated learning provides an efficient integrated model for distributed data, allowing the local training of different data. Meanwhile, the goal of multi-task learning is to simultaneously establish models for multiple related tasks, and to obtain the underlying main structure. However, traditional federated multi-task learning models not only have strict requirements for the data distribution, but also demand large amounts of calculation and have slow convergence, which hindered their promotion in many fields. In our work, we apply the rank constraint on weight vectors of the multi-task learning model to adaptively adjust the task's similarity learning, according to the distribution of federal node data. The proposed model has a general framework for solving optimal solutions, which can be used to deal with various data types. Experiments show that our model has achieved the best results in different dataset. Notably, our model can still obtain stable results in datasets with large distribution differences. In addition, compared with traditional federated multi-task learning models, our algorithm is able to converge on a local optimal solution within limited training iterations.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.04a
• /
• pp.633-635
• /
• 2016

Convolutional Networks (ConvNets) are powerful models that learn hierarchies of visual features, which could also be used to obtain image representations for transfer learning. The basic pipeline for transfer learning is to first train a ConvNet on a large dataset (source task) and then use feed-forward units activation of the trained ConvNet as image representation for smaller datasets (target task). Our key contribution is to demonstrate superior performance of multiple ConvNet layer features over single ConvNet layer features. Combining multiple ConvNet layer features will result in more complex feature space with some features being repetitive. This requires some form of feature selection. We use AdaBoost with single stumps to implicitly select only distinct features that are useful towards classification from concatenated ConvNet features. Experimental results show that using multiple ConvNet layer activation features instead of single ConvNet layer features consistently will produce superior performance. Improvements becomes significant as we increase the distance between source task and the target task.

• International conference on construction engineering and project management
• /
• 2020.12a
• /
• pp.387-396
• /
• 2020

Wearable biosensors have the potential to non-invasively and continuously monitor seniors' stress in their daily interaction with the urban environment, thereby enabling to address the stress and ultimately advance their outdoor mobility. However, current wearable biosensor-based stress detection methods have several drawbacks in field application due to their dependence on batch-learning algorithms. First, these methods train a single classifier, which might not account for multiple subjects' different physiological reactivity to stress. Second, they require a great deal of computational power to store and reuse all previous data for updating the signle classifier. To address this issue, we tested the feasibility of online multi-task learning (OMTL) algorithms to identify multiple seniors' stress from electrodermal activity (EDA) collected by a wristband-type biosensor in a daily trip setting. As a result, OMTL algorithms showed the higher test accuracy (75.7%, 76.2%, and 71.2%) than a batch-learning algorithm (64.8%). This finding demonstrates that the OMTL algorithms can strengthen the field applicability of the wearable biosensor-based stress detection, thereby contributing to better understanding the seniors' stress in the urban environment and ultimately advancing their mobility.

• International Journal of Precision Engineering and Manufacturing
• /
• v.7 no.1
• /
• pp.62-66
• /
• 2006

The learning control develops controllers that learn to improve their performance at executing a given task, based on experience performing this specific task. In a previous work, the authors presented an iterative precision of linear decentralized learning control based on p-integrated learning method for the vertical dynamic multiple systems. This paper develops an indirect decentralized learning control based on adaptive control method. The original motivation of the learning control field was learning in robots doing repetitive tasks such as an assembly line works. This paper starts with decentralized discrete time systems, and progresses to the robot application, modeling the robot as a time varying linear system in the neighborhood of the nominal trajectory, and using the usual robot controllers that are decentralized, treating each link as if it is independent of any coupling with other links. Some techniques will show up in the numerical simulation for vertical dynamic robot. The methods of learning system are shown for the iterative precision of each link.

• Korean Medical Education Review
• /
• v.15 no.1
• /
• pp.46-53
• /
• 2013

The purpose of this study was to analyze the relationship among medical students' learning motivation, characteristics of multiple intelligence, and academic achievement. The participants were 144 medical students. The data were collected by administering learning motivation tests (self-confidence, self-efficacy, level of task, emotion of learning, learning behavior, failure tolerance, task difficulty, and academic self-efficacy), a multiple intelligence test (linguistic intelligence, logical-mathematical intelligence, musical intelligence, bodily-kinesthetic intelligence, spatial intelligence, interpersonal intelligence, intrapersonal intelligence, and naturalistic intelligence), and two semesters of grades. There is a correlation between multiple intelligences and learning motivation. Among academic self-efficacy of academic motivation, the self-control efficacy (0.28) and behavior (0.18) subscales are significantly positively correlated with academic achievement. However, the emotion subscale (-0.18) was significantly negatively correlated. Learning motivation was correlated with two of the eight multiple intelligence profiles: the intrapersonal intelligence (0.18) and bodily-kinesthetic intelligence (-0.19). The structural equation modeling analysis showed that the behavior and self-control efficacy subscales of intrapersonal intelligence had an impact on academic achievement. An analysis according to the academic achievement group showed significant differences in self-control efficacy and emotion subscales with intrapersonal intelligence. A positive relationship can be observed between learning motivation and some characteristics of multiple intelligence of medical school students. In light of the findings, it is worth examining whether we can control medical students' learning motivation through educational programs targeting self-control efficacy and intrapersonal intelligence.

• Journal of the Korean Society for Precision Engineering
• /
• v.22 no.4
• /
• pp.92-98
• /
• 2005

The learning control develops controllers that learn to improve their performance at executing a given task, based on experience performing this specific task. In a previous work, the authors presented an iterative precision of linear decentralized learning control based on p-integrated learning method for the vertical dynamic multiple systems. This paper develops an indirect decentralized teaming control based on adaptive control method. The original motivation of the teaming control field was loaming in robots doing repetitive tasks such as on an assembly line. This paper starts with decentralized discrete time systems, and progresses to the robot application, modeling the robot as a time varying linear system in the neighborhood of the nominal trajectory, and using the usual robot controllers that are decentralized, treating each link as if it is independent of any coupling with other links. Some techniques will show up in the numerical simulation for vertical dynamic robot. The methods of learning system are shown up for the iterative precision of each link.

• Journal of Korea Society of Industrial Information Systems
• /
• v.1 no.1
• /
• pp.153-176
• /
• 1996

The new field of learning control devleops controllers that learn to improve their performance at executing a given task, based on experience performing this task. the simplest forms of learning control are based on the same concepts as integral control, but operating in the domain of the repetitions of the task. This paper studies the use of such controllers ina decentralized system, such as a robot with the controller for each link acting independently. The basic result of the paper is to show that stability of the learning controllers for all subsystems when the coupling between subsystems is turned off, assures stability of the decentralized learning in the coupled system, provided that the sample time in the digital learning controller is sufficiently short.

• Journal of Multimedia Information System
• /
• v.9 no.2
• /
• pp.75-86
• /
• 2022

The existing image quality assessment (IQA) datasets have a small number of samples. Some methods based on transfer learning or data augmentation cannot make good use of image quality-related features. A No Reference (NR)-IQA method based on multi-task training and quality awareness is proposed. First, single or multiple distortion types and levels are imposed on the original image, and different strategies are used to augment different types of distortion datasets. With the idea of weak supervision, we use the Full Reference (FR)-IQA methods to obtain the pseudo-score label of the generated image. Then, we combine the classification information of the distortion type, level, and the information of the image quality score. The ResNet50 network is trained in the pre-train stage on the augmented dataset to obtain more quality-aware pre-training weights. Finally, the fine-tuning stage training is performed on the target IQA dataset using the quality-aware weights to predicate the final prediction score. Various experiments designed on the synthetic distortions and authentic distortions datasets (LIVE, CSIQ, TID2013, LIVEC, KonIQ-10K) prove that the proposed method can utilize the image quality-related features better than the method using only single-task training. The extracted quality-aware features improve the accuracy of the model.

• Journal of Gifted/Talented Education
• /
• v.21 no.4
• /
• pp.961-977
• /
• 2011

The purpose of this study was to investigate the relationship between learning motivation and task commitment and find sub factors of learning motivation that affect task commitment. For this study 30 science gifted student (4th and 5th grade in elementary school) participated. The survey instruments used for this study were Academic Motivation Scale and Task Commitment Scale. The statistical methods employed for data analysis were the correlation analysis and multiple regression analysis. The result of this study were as follows: First, the learning motivation and task commitment of science gifted students showed similar levels. But there was differences of strength each sub factors of learning motivation and task commitment. Second, there was a significant positive correlation between learning motivation and task commitment. Also, learning motivation has the explanatory power of predictive variable for the task commitment approximately 49.3%. Expecially learning motivation has significant positive correlation with responsibility and self-control that sub factors of task commitment. Among the sub factor of learning motivation, confidence has most correlations with sub factors of task commitment and significant impact on task commitment. This result indicate that we need to develop learning motivation to improve task commitment and especially develop learning motivation program to grow up confidence of science-gifted.

• The Journal of Asian Finance, Economics and Business
• /
• v.5 no.2
• /
• pp.81-94
• /
• 2018

The purpose of this study is to substantiate the affecting factors of informal learning outcomes for professions in various dimensions of an individual and organization. In specific, the study analyzed the effects of learning motivation, job characteristics, and a supportive learning environment which have on task-related knowledge acquisition, adapting to organization and understanding contexts, relationship formation, and improving self-development-ability. The participants of the study were 261 professionals working at four major accounting firms in South Korea. Multiple regression models were applied step by step for analysis. In this study, the informal learning of professionals working at four major accounting firms is influenced by various factors of learning motivation, job characteristics, and a supportive learning environment. The detailed analysis results were as follows. Firstly, peer-support showed the most positive effect on task-related knowledge acquisition. Secondly, for adapting to organization and understanding contexts, task autonomy showed the greatest effect. Thirdly, peer-support was found to be the most important factor for relationship formation. Fourthly, for improving self-development ability, learning goal orientation showed to be the most important factor. The various factors facilitated the professional learning by empirical identification. The study presented practical implications for creating an effective informal learning support environment.