Adaptive Anomaly Detection for Internet of Things in Hierarchical Edge Computing: A Contextual-Bandit Approach

The advances in deep neural networks (DNN) have significantly enhanced real-time detection of anomalous data in IoT applications. However, the complexity-accuracy-delay dilemma persists: Complex DNN models offer higher accuracy, but typical IoT devices can barely afford the computation load, and the remedy of offloading the load to the cloud incurs long delay. In this article, we address this challenge by proposing an adaptive anomaly detection scheme with hierarchical edge computing (HEC). Specifically, we first construct multiple anomaly detection DNN models with increasing complexity and associate each of them to a corresponding HEC layer. Then, we design an adaptive model selection scheme that is formulated as a contextual-bandit problem and solved by using a reinforcement learning policy network . We also incorporate a parallelism policy training method to accelerate the training process by taking advantage of distributed models. We build an HEC testbed using real IoT devices and implement and evaluate our contextual-bandit approach with both univariate and multivariate IoT datasets. In comparison with both baseline and state-of-the-art schemes, our adaptive approach strikes the best accuracy-delay tradeoff on the univariate dataset and achieves the best accuracy and F1-score on the multivariate dataset with only negligibly longer delay than the best (but inflexible) scheme.

Dinâmicas Relacionais na Implementação de Políticas Públicas: Análise do Programa Minha Casa, Minha Vida – Entidades

Objetivo da Pesquisa: Investigar a implementação do Programa Minha Casa, Minha Vida – Entidades, em Conselheiro Lafaiete-MG, a partir das interações que emergem deste processo.Enquadramento Teórico: As perspectivas da Policy Network, considerada enquanto método e abordagem analítica, forneceram o quadro orientado à compreensão das dinâmicas relacionais.Metodologia: A coleta de dados foi realizada por meio de análises documentais e de entrevistas semiestruturadas, e o tratamento dos dados a partir da Análise de Conteúdo e da Análise de Redes Sociais.Resultados: Foi observado como as alterações nas naturezas das relações e a reconfigurações das redes foram condicionadas a aspectos contingenciais do ambiente no qual a política pública é implementada.Originalidade: O trabalho apresenta formas alternativas de emprego da Policy Network, sendo a abordagem relacional mista uma estratégia capaz de desvelar a dimensão política em um fenômeno que é subjetivo, variável e contextual.Contribuições Teóricas e Práticas: Evidenciou-se a necessidade de se compreender as dimensões históricas, político-institucionais e sociais que conformam a execução das políticas públicas. O entendimento desses elementos fornece subsídios para análises mais condizentes com a realidade, confirmando a fragilidade da compreensão da implementação como um simples output das etapas de agenda e formulação.

Professional Network Community for Training Youth Workers: Conceptual Framework and Prerequisites for the Creation

Currently, in the context of the transition to hybrid education, new network forms of communication and interaction of teachers, students as well as specialists are in demand in the framework of personnel training, research and professional activities. The article reveals the main aspects of the integrated network community for training professionals working with youth in the field of educational and youth social policy. Network community can be viewed as an important resource for the development of both the participants themselves and the specific professional sphere. The purpose of the article is to describe the conceptual framework of a professional network community for training personnel for working with youth in the context of social, educational and youth policy and identify the prerequisites for its creation.In the course of the work, we used such methods as structural and functional analysis, system synthesis, modeling and an online survey. The sample comprised university students, faculty members, and youth workers (n = 147) from six federal districts of Russia.The conceptual framework of the professional network community presented on the basis of systemic-synergetic and constructive methodological approaches enables 1) to formulate its idea as a collaboration for personnel training, scientific research, joint projects implementation and self-development of participants; 2) to identify systemic contradictions, the solution of which is directed by the interaction of community members; 3) to determine the axiological basis, purpose, objectives and stages of deployment; 4) to identify the degree of its relevance, the presence of common interests among all its participants on the basis of the results of empirical research that may be useful in the professional network communities design.

Robust Walking Control of a Lower Limb Rehabilitation Exoskeleton Coupled with a Musculoskeletal Model via Deep Reinforcement Learning

Background: Few studies have systematically investigated robust controllers for lower limb rehabilitation exoskeletons (LLREs) that can safely and effectively assist users with a variety of neuromuscular disorders to walk with full autonomy. One of the key challenges for developing such a robust controller is to handle different degrees of uncertain human-exoskeleton interaction forces from the patients. Consequently, conventional walking controllers either are patient-condition specific or involve tuning of many control parameters, which could behave unreliably and even fail to maintain balance. Methods: We present a novel and robust controller for a LLRE based on a decoupled deep reinforcement learning framework with three independent networks, which aims to provide reliable walking assistance against various and uncertain human-exoskeleton interaction forces. The exoskeleton controller is driven by a neural network control policy that acts on a stream of the LLRE’s proprioceptive signals, including joint kinematic states, and subsequently predicts real-time position control targets for the actuated joints. To handle uncertain human-interaction forces, the control policy is trained intentionally with an integrated human musculoskeletal model and realistic human-exoskeleton interaction forces. Two other neural networks are connected with the control policy network to predict the interaction forces and muscle coordination. To further increase the robustness of the control policy, we employ domain randomization during training that includes not only randomization of exoskeleton dynamics properties but, more importantly, randomization of human muscle strength to simulate the variability of the patient’s disability. Through this decoupled deep reinforcement learning framework, the trained controller of LLREs is able to provide reliable walking assistance to the human with different degrees of neuromuscular disorders. Results and Conclusion: A universal, RL-based walking controller is trained and virtually tested on a LLRE system to verify its effectiveness and robustness in assisting users with different disabilities such as passive muscles (quadriplegic), muscle weakness, or hemiplegic conditions. An ablation study demonstrates strong robustness of the control policy under large exoskeleton dynamic property ranges and various human-exoskeleton interaction forces. The decoupled network structure allows us to isolate the LLRE control policy network for testing and sim-to-real transfer since it uses only proprioception information of the LLRE (joint sensory state) as the input. Furthermore, the controller is shown to be able to handle different patient conditions without the need for patient-specific control parameters tuning.

A Reinforcement Learning Based Dirt-Exploration for Cleaning-Auditing Robot

Cleaning is one of the fundamental tasks with prime importance given in our day-to-day life. Moreover, the importance of cleaning drives the research efforts towards bringing leading edge technologies, including robotics, into the cleaning domain. However, an effective method to assess the quality of cleaning is an equally important research problem to be addressed. The primary footstep towards addressing the fundamental question of “How clean is clean” is addressed using an autonomous cleaning-auditing robot that audits the cleanliness of a given area. This research work focuses on a novel reinforcement learning-based experience-driven dirt exploration strategy for a cleaning-auditing robot. The proposed approach uses proximal policy approximation (PPO) based on-policy learning method to generate waypoints and sampling decisions to explore the probable dirt accumulation regions in a given area. The policy network is trained in multiple environments with simulated dirt patterns. Experiment trials have been conducted to validate the trained policy in both simulated and real-world environments using an in-house developed cleaning audit robot called BELUGA.

Education in times of global crisis: How private actors in Germany gain power through Twitter

The outbreak of the COVID-19 pandemic led to enormous societal changes worldwide and touched many different areas of daily life. One of the most serious restrictions to contain the pandemic was the closure of schools and kindergartens. Particularly in countries with comparatively low levels of digitalization in schools, this situation opened up opportunities for private actors to gain importance and influence in the education sector. For this article, we draw on policy network approaches and network theory to analyze Twitter discussions around digital learning and homeschooling during the period of school closures in Germany due to the COVID-19 crisis. We use social network analysis to identify the actors involved and their influences in the issue-specific Twitter communication network. In recent years, Twitter has been increasingly used for exchanges on education policy content, mainly by political and civil society actors. Our study shows that with respect to digital learning and homeschooling, it was primarily individual experts and consultants as well as corporations which influenced the discourse. In particular, it appears that Twitter is used as a forum to promote corporations’ own products and platforms, including by globally operating corporations such as Microsoft and YouTube, while public actors remain barely visible.

District health network policy in Iran: the role of ideas, interests, and institutions (3i framework) in a nutshell

Background District Health Network (DHN), one of Iran’s most successful health reforms, was launched in 1985 to provide primary health care (PHC), in response to health inequities in Iran. The present study aims to use interrelated elements of the 3i framework: ideas (e.g., beliefs and values, culture, knowledge, research evidence and solutions), interests (e.g., civil servants, pressure groups, elected parties, academians and researchers, and policy entrepreneurs), and institutions (e.g., rules, precedents, and organizational, government structures, policy network, and policy legacies) to explain retrospectively how (DHN) policy in Iran, as a developing country, was initiated and formed. Methods A historical narrative approach with a case study perspective was employed to focus on the formation and framing process of DHN. For this purpose, the 3i framework was used as a guideline for data analysis. This study mainly searched and extracted secondary sources, including online news, reports, books, dissertations, and published articles in the scientific databases. Primary interviews as a supplementary source were also carried out to meet cross-validation of the data. Data were analyzed using a deductive and inductive approach. Results According to the 3i framework, the following factors contributed to the formation of DHN policy in Iran: previous national efforts (for instance Rezaieh plan) and international events aiming to provide public health services for peripheral regions; dominant social discourses and values at the beginning of the Iranian revolution such as addressing the needs of disadvantaged and marginalized groups, which were embedded in the goals of DHN policy aiming to provide basic health services for deprived people especially living in rural and remote areas. Besides, the remarkable social cohesion and solidarity among people reinforced by the Iran-Iraq war were among other factors which contributed to the formation of participatory plans such as DHN (ideas). Main policy entrepreneurs including Minister of Health, his public health deputy and two planners of DHN with similar and rich background in the public health field and sharing the same beliefs (interests) which subsequently led to creation of tight-knit policy community network between them (institutions) also accelerated the creation of DHN in Iran to great extent. Political support of parliamentary representatives (interests), and formal laws such as principles of Iran Constitution (institutions) were also influential in passing the DHN in Iran. Conclusions The 3i framework constituents would be insightful in explaining the creation of public health policies. This framework showed that the alignment of laws, structures, and interests of the main actors of the policy with the dominant ideas and beliefs in the society, opened the opportunity to form DHN in Iran.