Mobile edge computing (MEC) is a promising solution to support resource-constrained devices by offloading tasks to the edge servers. However, traditional approaches (e.g., linear programming and ...game-theory methods) for computation offloading mainly focus on the immediate performance, potentially leading to performance degradation in the long run. Recent breakthroughs regarding deep reinforcement learning (DRL) provide alternative methods, which focus on maximizing the cumulative reward. Nonetheless, there exists a large gap to deploy real DRL applications in MEC. This is because: 1) training a well-performed DRL agent typically requires data with large quantities and high diversity, and 2) DRL training is usually accompanied by huge costs caused by trial-and-error. To address this mismatch, we study the applications of DRL on the multi-user computation offloading problem from a more practical perspective. In particular, we propose a distributed and collective DRL algorithm called DC-DRL with several improvements: 1) a distributed and collective training scheme that assimilates knowledge from multiple MEC environments, which not only greatly increases data amount and diversity but also spreads the exploration costs, 2) an updating method called adaptive n-step learning, which can improve training efficiency without suffering from the high variance caused by distributed training, and 3) combining the advantages of deep neuroevolution and policy gradient to maximize the utilization of multiple environments and prevent the premature convergence. Lastly, evaluation results demonstrate the effectiveness of our proposed algorithm. Compared with the baselines, the exploration costs and final system costs are reduced by at least 43 and 9.4 percent, respectively.
One of the psychological mechanisms that contribute to effective and efficient team actions is team cognition, defined either as shared knowledge states about game situations, teammates' skills, and ...action probabilities or direct communication processes in the team action itself. Particularly in interactive team sports (e.g., football), characterized by highly complex, dynamic, and uncertain situations, sharing a common understanding concerning potential future actions and how to coordinate these actions may be an advantage. Otherwise, team members must communicate their thoughts and ideas on the fly, which might be impossible due to time pressure, cognitive costs or noisy environments. This study examined if shared knowledge and verbal communication change through collective training. Forty-six under-18 and under-21 youth football players performed a football task in teams of two. The task consisted of passing and running elements common in football. After a training phase, and before two testing phases, players evaluated their actions and the actions of their assigned teammate regarding action type, location, and timing. Out of these evaluations, two indices of common understanding were computed. Furthermore, verbal communication during the task was video-and audio-recorded. Data analysis showed that shared knowledge considerably increased over time and with practice. Simultaneously, overall verbal communication and verbal communication consisting of orienting information was significantly reduced. Additionally, there was a tendency for a correlation that when shared knowledge increased, orienting verbal communication decreased. Overall, the players used orienting communications the most (77%). The study revealed that shared knowledge states and verbal communication change through collective training and that there might be a relation between the level of shared knowledge and the use of orienting verbal communication. Further studies in and off the field are needed to disentangle the complex interplay of team cognitions.
Medical personnel in the UK Armed Forces are highly trained to deploy in support of military operations that assist humanitarian, peacekeeping, counter-terrorism and environmental catastrophes ...anywhere in the world. Such environments are often austere and successful outcomes demand an individual is highly resilient and able to adapt quickly to any situation. This qualitative study aimed to determine the factors that affect healthcare delivery on such missions by capturing the personal experiences of the first military personnel deployed on a humanitarian operation in support of the Ebola outbreak in West Africa between October 2014 and January 2015.
A grounded theory methodology was utilised to probe the personal accounts of these experiences. Semi-structured interviews were conducted on 14 multi-disciplinary personnel 3–6 months following their return to the UK and were transcribed verbatim. Data were analysed and a framework generated that had been further refined by discussion with military personnel independent of the study but with the contextual understanding and experience of this particular deployment.
The resultant theoretical framework was underpinned by participants framing their experience by “just getting on with it”. Stressors such as a poor flow of information, a fear of the unknown, strict patient admission criteria, environmental constraints and transcultural boundaries to care were mitigated by strong leadership, teamwork, peer support and the positive impact of having made a difference.
Collective pre-deployment training generated competence, confidence and team cohesiveness providing a firm foundation for coping with the challenges of this humanitarian mission, which continued to be strengthened throughout the deployment. These factors helped to build personnel's resilience to the stressors associated with the mission and may help to protect their mental health outcomes in the longer-term.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK, ZRSKP
The goal of this research was the development of a practical architecture for the computer-based tutoring of teams. This article examines the relationship of team behaviors as antecedents to ...successful team performance and learning during adaptive instruction guided by Intelligent Tutoring Systems (ITSs). Adaptive instruction is a training or educational experience tailored by artificially-intelligent, computer-based tutors with the goal of optimizing learner outcomes (e.g., knowledge and skill acquisition, performance, enhanced retention, accelerated learning, or transfer of skills from instructional environments to work environments). The core contribution of this research was the identification of behavioral markers associated with the antecedents of team performance and learning thus enabling the development and refinement of teamwork models in ITS architectures. Teamwork focuses on the coordination, cooperation, and communication among individuals to achieve a shared goal. For ITSs to optimally tailor team instruction, tutors must have key insights about both the team and the learners on that team. To aid the modeling of teams, we examined the literature to evaluate the relationship of teamwork behaviors (e.g., communication, cooperation, coordination, cognition, leadership/coaching, and conflict) with team outcomes (learning, performance, satisfaction, and viability) as part of a large-scale meta-analysis of the ITS, team training, and team performance literature. While ITSs have been used infrequently to instruct teams, the goal of this meta-analysis make team tutoring more ubiquitous by: identifying significant relationships between team behaviors and effective performance and learning outcomes; developing instructional guidelines for team tutoring based on these relationships; and applying these team tutoring guidelines to the Generalized Intelligent Framework for Tutoring (GIFT), an open source architecture for authoring, delivering, managing, and evaluating adaptive instructional tools and methods. In doing this, we have designed a domain-independent framework for the adaptive instruction of teams.
Full text
Available for:
EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NLZOH, NUK, OBVAL, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
This study proposes a competitive intelligence connectivist Massive Open Online Course (CI cMOOC) proof of concept and highlights the interactions among content, context and community to explore ...relevance in CI cMOOC behavior. The CI cMOOC proof of concept was empirically tested with an online purposive sampling to target a qualified audience of similar and dissimilar information-rich cases, providing evidence about content-context-community competing influence on CI knowledge. The results revealed how the CI learning community perceive the capability of a cMOOC to train foreknowledge practices, given the best match between its content and context. The findings outline that tailored learning approach of the instructor influences the CI learning community’s satisfaction with the content. The study facilitates theory development in addressing the emerging paradigm of an open intelligence approach to cMOOC collective training. Within boundaries of empirical return on experience of qualified respondents, the research framework strengthens trust in supervised interpretive judgment of CI learners confronted with anticipating competitive challenges.
Total laryngectomy affects the speaking functions of many patients. Speech deprivation has great impacts on the quality of life of patients, especially on self-efficacy. Learning esophageal speech ...represents a way to help laryngectomees speak again. The purpose of this study was to determine the influence of collective esophageal speech training on self-efficacy of laryngectomees. In this study, 28 patients and 30 family members were included. The participants received information about training via telephone or a WeChat group. Collective esophageal speech training was used to educate laryngectomees on esophageal speech. Before and after collective esophageal speech training, all participants completed the General Self-Efficacy Scale (GSES) to assess their perceptions on self-efficacy. Through the training, laryngectomees recovered their speech. After the training, the self-efficacy scores of laryngectomees were higher than those before the training, with significant differences noted (T<0.05). However, family members' scores did not change significantly. In conclusion, collective esophageal speech training is not only convenient and economical, but also improves self-efficacy and confidence of laryngectomees. Greater self-efficacy is helpful for laryngectomees to master esophageal speech and improve their quality of life. In addition, more attention should be focused on improving the self-efficacy of family members and making them give full play to their talent and potential on laryngectomees' voice rehabilitation.
Full text
Available for:
EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NLZOH, NUK, OBVAL, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
This paper presents the findings of a United Kingdom (UK) research program carried out over the last decade. This research has explored the benefits of using networks of simulators for collective ...training known in the UK as mission training through distributed simulation (MTDS). The paper provides an overview of trials carried out to date, identifies the research issues addressed, and discusses the key findings. The conclusion is that MTDS provides an immersive training environment that has the potential to support not only single service collective training, but also joint and coalition training requirements.
Full text
Available for:
BFBNIB, NUK, PILJ, SAZU, UL, UM, UPUK
10.
Dual-Guided Collective Entity Alignment for Knowledge Graphs Bai, Gang; Kou, Yue; Shen, Derong ...
2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys),
2022-Dec.
Conference Proceeding
Entity alignment aims to link entities representing the same thing in different knowledge graphs, which is the primary task of knowledge graph fusion. Most of the existing methods only rely on the ...local structure of entities and ignore the difference of entities' associated structure in different knowledge graphs. So it is difficult for them to align long-tail entities with sparse local structures. In this paper, we propose a Dual-Guided Collective Entity Alignment Model (DGC-Align) for knowledge graphs, which comprehensively considers two kinds of associated structures-associated path and local structure. We use meta-paths and neighborhoods as guidance respectively to mine the two associated structure and make adaptive integration, and add the relation information to distinguish different types of information. It can effectively alleviate the problem of long-tail entity alignment caused by the sparse knowledge structure, and has good adaptability for the alignment of entities with different associated structures in heterogeneous graphs. The feasibility and effectiveness of the model we proposed are verified by experiments.
PDF
