Academic literature on the topic 'Beast Fables'

This paper proposes a novel direct policy search (DPS) method with model selection for partially observed Markov decision processes (POMDPs). DPSs have been standard for learning POMDPs due to their computational efficiency and natural ability to maximize total rewards. An important open challenge for the best use of DPS methods is model selection, i.e., determination of the proper dimensionality of hidden states and complexity of policy functions, to mitigate overfitting in highly-flexible model representations of POMDPs. This paper bridges Bayesian inference and reward maximization and derives marginalized weighted log-likelihood~(MWL) for POMDPs which takes both advantages of Bayesian model selection and DPS. Then we propose factorized asymptotic Bayesian policy search (FABPS) to explore the model and the policy which maximizes MWL by expanding recently-developed factorized asymptotic Bayesian inference. Experimental results show that FABPS outperforms state-of-the-art model selection methods for POMDPs, with respect both to model selection and to expected total rewards.

In aero engines, the rotating shafts are supported by a set of bearings, which are enclosed in bearing chambers. Cooling and lubrication oil escapes from the bearings and these chambers are designed to capture and recycle it. A good understanding of the oil behaviour inside bearing chambers is therefore desirable in order to limit the oil volume involved and minimize transmission losses whilst managing the engine core heat in the best possible manner. This study is focused on the simulation of the oil behaviour inside such a chamber and special attention is given to the so-called KIT bearing chamber. The oil phase in the chamber can take different forms e.g. sprays, droplets, thin films or a combination of those. Assuming the oil we want to track remains dominantly as a film and large droplets/filaments, the Volume of Fluid (VOF) method is used in order to track the oil and the oil/air interface in the chamber, hereby investigating the feasibility and merits of such an approach and extending the earlier work carried out by the authors and colleagues. An Enhanced VOF approach coupled with level-set is used here unless stated otherwise. The simulated pump outlet condition, proposed by the University of Nottingham, is also employed in this study, to replicate an engine displacement pump. Since the use of VOF requires a refined mesh in the oil region, an adaptive mesh approach based on the volume of fluid gradient is developed and validated to control the total cell count for some of the cases reported here and limit simulation costs. The Adaptive Mesh Approach (AMA) can allow a better resolution of critical interfaces, better compute the oil break-up (within the limitation of the physical models used) and then track the droplets and filaments. Therefore, not only the CPU time cost might be reduced compared to a fixed mesh approach but significant physical aspects of the problem should be better accounted for. In order to inform the set up and parameters used with this method, and appraise its value for the proposed application, the experimental study of Fabre is used before the approach is applied to the KIT chamber. Good insight is obtained in terms of run time acceleration for such problem when combining the proposed VOF setup with adaptive meshing. Key set up parameters are quantified. The simulations carried out with the proposed set up are proving to be fairly robust and stable. Qualitative (physical) evidence is also encouraging and confirms the value of such an approach to the study of aeroengine bearing chambers.

With the popularity and spread of social media, more and more social software is helping to bring people closer to each other [1]. It is increasingly easy for adolescents to get other people's updates from social media, including celebrities, internet celebrities and peers [2]. Also adolescence is a time when the brain undergoes many structural and functional changes, so it is likely that the part of the social brain responsible for regulating imitation is still maturing throughout adolescence, which may lead to more pronounced imitative behaviour [3]. In addition, adolescents gain popularity, status and attractiveness through imitation of their idols or among their peers [4]. Therefore, making good use of the prevalent behaviours that social media has created in society has the potential to provide better behavioural interventions for the adolescent population [5], helping to shape better behavioural habits in adolescents, improving the current trend of younger disease and potentially reducing the likelihood of preventable health problems.The aim of this study was to analyse how popular behavioural mimicry among adolescents can be used to promote the activation of their health behaviours. We asked two questions: 1. the extent to which imitation behaviours activate adolescents' health behaviours; 2. measuring the impact of knowledge, skills and beliefs involved in the activation of behavioural imitation on adolescents' health maintenance and disease prevention.A questionnaire was used to enumerate the population groups that have the greatest influence on adolescents as the test sample in this study. 100 participants took part in the questionnaire, including 50 participants from mainland China and 50 participants from Hong Kong, whose mean age was 16 ± 3 years. After administering the questionnaire, 50 of these participants, who were randomly and equally divided into 10 groups of 5 participants each, were surveyed using the Activation Inventory (PAM) to measure the current level of knowledge, skills and beliefs involved in the activation of the adolescent population to maintain health and prevent disease, and then measured again using the PAM 30 and 60 days after the adolescents were exposed to the imitated subjects.The adolescent group itself was not highly aware of healthy behaviours and the effectiveness of positive health behaviour imitation in changing health behaviours and outcomes was somewhat proven when they were exposed to positive health behaviours of imitators for 30 days. However, 60 days after participants were exposed to imitations of healthy behaviours, although the imitations were still effective in maintaining healthy behaviours, the 60-day activation of healthy behaviours produced some decline compared to the first 30 days of outcomes. Therefore, in the future, more research should be conducted on the preferences and needs of adolescent groups to identify the social factors and groups that best trigger imitation among adolescents, and to promote positive health behaviours among adolescents by developing mobile applications that are more in line with adolescents' expectations to trigger trends, create widespread social discussion and be present in their daily conversations.References1.Moira Burke and Robert E. Kraut. 2014. Growing closer on facebook: changes in tie strength through social network site use. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '14). Association for Computing Machinery, New York, NY, USA, 4187–4196. https://doi.org/10.1145/2556288.25570942.C. Longobardi, M. Settanni, M.A. Fabris, D. Marengo, Follow or be followed: Exploring the links between Instagram popularity, social media addiction, cyber victimization, and subjective happiness in Italian adolescents, Children and Youth Services Review, Volume 113, 2020, 104955,ISSN 0190-7409, https://doi.org/10.1016/j.childyouth.2020.104955.3.Cook, J., Bird, G. Social attitudes differentially modulate imitation in adolescents and adults. Exp Brain Res 211, 601–612 (2011). https://doi.org/10.1007/s00221-011-2584-4.4.Raviv, A., Bar-Tal, D., Raviv, A. et al. Adolescent idolization of pop singers: Causes, expressions, and reliance. J Youth Adolescence 25, 631–650 (1996). https://doi.org/10.1007/BF01537358.5.Korda H, Itani Z. Harnessing Social Media for Health Promotion and Behavior Change. Health Promotion Practice. 2013;14(1):15-23. https://doi.org/10.1177/1524839911405850.