• ETRI Journal
• /
• v.43 no.6
• /
• pp.1004-1012
• /
• 2021

The collaboration productively interacting between multi-agents has become an emerging issue in real-world applications. In reinforcement learning, multi-agent environments present challenges beyond tractable issues in single-agent settings. This collaborative environment has the following highly complex attributes: sparse rewards for task completion, limited communications between each other, and only partial observations. In particular, adjustments in an agent's action policy result in a nonstationary environment from the other agent's perspective, which causes high variance in the learned policies and prevents the direct use of reinforcement learning approaches. Unexpected social loafing caused by high dispersion makes it difficult for all agents to succeed in collaborative tasks. Therefore, we address a paradox caused by the social loafing to significantly reduce total returns after a certain timestep of multi-agent reinforcement learning. We further demonstrate that the collaborative paradox in multi-agent environments can be avoided by our proposed effective early stop method leveraging a metric for social loafing.

• Journal of muscle and joint health
• /
• v.25 no.3
• /
• pp.240-249
• /
• 2018

Purpose: This study was to identify the attributes, antecedents, their consequences, and empirical indicators of rehabilitation motivation in rheumatoid arthritis patients. Methods: Walker and Avant's method was used to analyze the concept. Articles published after 1990 were searched in Medline, CINAHL, NSDL, and RISS databases using "rehabilitation", "motivation" and their combination as keywords. Results: The attributes of rehabilitation motivation are: 1) certitude and trust toward rehabilitation treatment; 2) confidence in the rehabilitation process; 3) efforts and commitments to achieve health goals; 4) psychological needs to act toward health recovery. Its antecedents include: 1) rights of self-determination; 2) goal setting and goal-oriented attitude; 3) personal needs; 4) getting rewards; 5) social and family support; 6) professional behavior of healthcare providers; and 7) least risks or costs for actions taken. Conclusion: The study results could be used as a conceptual framework for developing tools to measure the motivation of rheumatoid arthritis patients.

• Nuclear Engineering and Technology
• /
• v.54 no.9
• /
• pp.3283-3292
• /
• 2022

Based on the Deep Q-Network(DQN) algorithm of reinforcement learning, an active fault-tolerance method with incremental action is proposed for the control system with sensor faults of the once-through steam generator(OTSG). In this paper, we first establish the OTSG model as the interaction environment for the agent of reinforcement learning. The reinforcement learning agent chooses an action according to the system state obtained by the pressure sensor, the incremental action can gradually approach the optimal strategy for the current fault, and then the agent updates the network by different rewards obtained in the interaction process. In this way, we can transform the active fault tolerant control process of the OTSG to the reinforcement learning agent's decision-making process. The comparison experiments compared with the traditional reinforcement learning algorithm(RL) with fixed strategies show that the active fault-tolerant controller designed in this paper can accurately and rapidly control under sensor faults so that the pressure of the OTSG can be stabilized near the set-point value, and the OTSG can run normally and stably.

• Asia Marketing Journal
• /
• v.19 no.3
• /
• pp.45-74
• /
• 2017

The current research proposes to fill a research gap by testing how reward uncertainty, different types of motivation, as well as individual risk-taking attitude affect consumers' promotion participation. Being offered with an uncertain reward, relative to individuals with extrinsic motivation, individuals with intrinsic motivation will have greater intention to participate in marketing promotion. In contrast, being offered with a certain reward, relative to individuals with intrinsic motivation, individuals with extrinsic motivation will have greater intention to participate in marketing promotion. This effect arises only among consumers having a low level of risk-taking attitude. For consumers having a high level of risk-taking attitude, their participation intention shows no significant difference between the two motivation type groups, under both certain and uncertain reward conditions. With an understanding of how consumer's response heterogeneously to promotions involving rewards, marketers can better understand not only how to use this promotional tactic more effectively, but also how to better allocate their budget for promotions.

• Asia Marketing Journal
• /
• v.18 no.4
• /
• pp.77-98
• /
• 2017

This research explores two key facets of behavior (reward point accrual and redemption) that consist of a loyalty program. It focuses on assessing the impact of accrual effort level on three types of redemption behavior: speed, unit size, and hedonic preference at the individual level by using large scale transaction data from a multi-vendor loyalty program providing flexible environment for point accrual and redemption. Findings from this research demonstrate that customers tend 1) to speed up point redemption, 2) to enlarge the size of redeemed points, and 3) to prefer utilitarian rewards as the level of effort at the accrual stage of reward point increases.

• International conference on construction engineering and project management
• /
• 2017.10a
• /
• pp.245-251
• /
• 2017

Integrated Project Delivery (IPD) is a procurement method that has been proved to improve construction project performance. However, in China implementation of IPD practices in construction projects is unknown though some researchers have studied the problems and constraints in adoption IPD. The purpose of this study was to explore IPD adoption in Chinese construction industry. Critical components of IPD implementation were reviewed, and questionnaires were distributed to collect industry views. The results revealed that IPD uptake is still low. In particular, the liability waiver and shared risks and rewards have been rarely used. In addition, co-location, value engineering method and the new compensation approach have also been hardly adopted. Some practices related to early involvement of key parties were adopted. Surprisingly, the findings indicate that the client has been continuously involved in the projects. The findings may imply that the legal issues and problems of contractual frameworks are still constraining IPD implementation in Chinese construction industry.

• ETRI Journal
• /
• v.45 no.5
• /
• pp.735-745
• /
• 2023

This paper surveys recent multiagent reinforcement learning and neural Myerson auction deep learning efforts to improve mobility control and resource management in autonomous ground and aerial vehicles. The multiagent reinforcement learning communication network (CommNet) was introduced to enable multiple agents to perform actions in a distributed manner to achieve shared goals by training all agents' states and actions in a single neural network. Additionally, the Myerson auction method guarantees trustworthiness among multiple agents to optimize rewards in highly dynamic systems. Our findings suggest that the integration of MARL CommNet and Myerson techniques is very much needed for improved efficiency and trustworthiness.

• Asia pacific journal of information systems
• /
• v.30 no.1
• /
• pp.143-165
• /
• 2020

The rapid growth of economic transactions generated by live streaming broadcasts ("LSB") has created opportunities for retailers to increase sales. However, little is known about what impact LSB celebrities have on customers and what causes LSB celebrities to become famous. This study aimed to fill this gap by studying the economics of LSBs. This study was conducted through a para-social relationship and attractiveness theory framework. Consequently, social and task attraction were assumed to be the antecedents of the para-social relationship that induced purchase intention. This study examined the impact of relationship rewards, self-disclosure, affective interactivity, informative interactivity, and the amount of information provided on purchase intentions through LSB. Celebrities can use the results of this study to enhance their appeal to fans and promote customers' purchase on e-commerce. This study contributed to the IS field by investigate the impact of para-social relationship on the online shopping context.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.192-198
• /
• 2024

This paper investigates the application of multi-agent deep reinforcement learning in the fighting game Samurai Shodown using Proximal Policy Optimization (PPO) and Advantage Actor-Critic (A2C) algorithms. Initially, agents are trained separately for 200,000 timesteps using Convolutional Neural Network (CNN) and Multi-Layer Perceptron (MLP) with LSTM networks. PPO demonstrates superior performance early on with stable policy updates, while A2C shows better adaptation and higher rewards over extended training periods, culminating in A2C outperforming PPO after 1,000,000 timesteps. These findings highlight PPO's effectiveness for short-term training and A2C's advantages in long-term learning scenarios, emphasizing the importance of algorithm selection based on training duration and task complexity. The code can be found in this link https://github.com/Lexer04/Samurai-Shodown-with-Reinforcement-Learning-PPO.

• The Journal of Society for e-Business Studies
• /
• v.20 no.1
• /
• pp.45-59
• /
• 2015

As many companies have been interested in big data, they have invested a lot of resources to get more customer data. Some companies try to trade the data illegally. In order to collect more customer data, companies provide various incentive programs to customers. However, their results are normally much less than their expectations. This study focuses on exploring the relative importance of the factors which influence customer attitudes to providing his/her personal information. This study conducts a conjoint analysis to assess trade-offs among the five influential factors-monetary reward, concern for data collection, concern for secondary use, concern for unauthorized use, and concern for errors. This study finds that the customer attitude to providing personal information is most influenced by the concern for secondary use. Furthermore, it shows that there are some differences between the light internet user group and the heavy internet user group in the relative importances of these factors. The monetary rewards appeal to the heavy internet users, rather than the light internet users.