RESUMO
The widespread adoption of cloud computing necessitates privacy-preserving techniques that allow information to be processed without disclosure. This paper proposes a method to increase the accuracy and performance of privacy-preserving Convolutional Neural Networks with Homomorphic Encryption (CNN-HE) by Self-Learning Activation Functions (SLAF). SLAFs are polynomials with trainable coefficients updated during training, together with synaptic weights, for each polynomial independently to learn task-specific and CNN-specific features. We theoretically prove its feasibility to approximate any continuous activation function to the desired error as a function of the SLAF degree. Two CNN-HE models are proposed: CNN-HE-SLAF and CNN-HE-SLAF-R. In the first model, all activation functions are replaced by SLAFs, and CNN is trained to find weights and coefficients. In the second one, CNN is trained with the original activation, then weights are fixed, activation is substituted by SLAF, and CNN is shortly re-trained to adapt SLAF coefficients. We show that such self-learning can achieve the same accuracy 99.38% as a non-polynomial ReLU over non-homomorphic CNNs and lead to an increase in accuracy (99.21%) and higher performance (6.26 times faster) than the state-of-the-art CNN-HE CryptoNets on the MNIST optical character recognition benchmark dataset.
Assuntos
Segurança Computacional , Redes Neurais de Computação , Privacidade , Humanos , Algoritmos , Computação em NuvemRESUMO
This manuscript describes the development of a resources module that is part of a learning platform named 'NIGMS Sandbox for Cloud-based Learning' https://github.com/NIGMS/NIGMS-Sandbox. The overall genesis of the Sandbox is described in the editorial NIGMS Sandbox at the beginning of this Supplement. This module delivers learning materials on implementing deep learning algorithms for biomedical image data in an interactive format that uses appropriate cloud resources for data access and analyses. Biomedical-related datasets are widely used in both research and clinical settings, but the ability for professionally trained clinicians and researchers to interpret datasets becomes difficult as the size and breadth of these datasets increases. Artificial intelligence, and specifically deep learning neural networks, have recently become an important tool in novel biomedical research. However, use is limited due to their computational requirements and confusion regarding different neural network architectures. The goal of this learning module is to introduce types of deep learning neural networks and cover practices that are commonly used in biomedical research. This module is subdivided into four submodules that cover classification, augmentation, segmentation and regression. Each complementary submodule was written on the Google Cloud Platform and contains detailed code and explanations, as well as quizzes and challenges to facilitate user training. Overall, the goal of this learning module is to enable users to identify and integrate the correct type of neural network with their data while highlighting the ease-of-use of cloud computing for implementing neural networks. This manuscript describes the development of a resource module that is part of a learning platform named ``NIGMS Sandbox for Cloud-based Learning'' https://github.com/NIGMS/NIGMS-Sandbox. The overall genesis of the Sandbox is described in the editorial NIGMS Sandbox [1] at the beginning of this Supplement. This module delivers learning materials on the analysis of bulk and single-cell ATAC-seq data in an interactive format that uses appropriate cloud resources for data access and analyses.
Assuntos
Aprendizado Profundo , Redes Neurais de Computação , Humanos , Pesquisa Biomédica , Algoritmos , Computação em NuvemRESUMO
This work investigated the annual variations in dry snow (DSRZ) and wet snow radar zones (WSRZ) in the north of the Antarctic Peninsula between 2015-2023. A specific code for snow zone detection on Sentinel-1 images was created on Google Earth Engine by combining the CryoSat-2 digital elevation model and air temperature data from ERA5. Regions with backscatter coefficients (σ°) values exceeding -6.5 dB were considered the extent of surface melt occurrence, and the dry snow line was considered to coincide with the -11 °C isotherm of the average annual air temperature. The annual variation in WSRZ exhibited moderate correlations with annual average air temperature, total precipitation, and the sum of annual degree-days. However, statistical tests indicated low determination coefficients and no significant trend values in DSRZ behavior with atmospheric variables. The results of reducing DSRZ area for 2019/2020 and 2020/2021 compared to 2018/2018 indicated the upward in dry zone line in this AP region. The methodology demonstrated its efficacy for both quantitative and qualitative analyses of data obtained in digital processing environments, allowing for the large-scale spatial and temporal variations monitoring and for the understanding changes in glacier mass loss.
Assuntos
Computação em Nuvem , Radar , Neve , Regiões Antárticas , Estações do Ano , Monitoramento Ambiental/métodos , TemperaturaRESUMO
SUMMARY: Volume abnormalities in subcortical structures, including the hippocampus, amygdala, thalamus, caudate, putamen, and globus pallidus have been observed in schizophrenia (SZ) and bipolar disorder (BD), not all individuals with these disorders exhibit such changes. In addition, the specific patterns and severity of volume changes may vary between individuals and at different stages of the disease. The study aims to compare the volumes of these subcortical structures between healthy subjects and individuals diagnosed with SZ or BD. Volumetric measurements of lateral ventricle, globus palllidus, caudate, putamen, hippocampus, and amygdale were made by MRI in 52 healthy subjects (HS), 33 patients with SZ, and 46 patients with BD. Automatic segmentation methods were used to analyze the MR images with VolBrain and MRICloud. Hippocampus, amygdala and lateral ventricle increased in schizophrenia and bipolar disorder patients in comparison with control subjects using MRIcloud. Globus pallidus and caudate volume increased in patients with schizophrenia and bipolar disorder compared control subjects using Volbrain. We suggested that our results will contribute in schizophrenia and bipolar disorder patients that assessment of the sub-cortical progression, pathology, and anomalies of subcortical brain compositions. In patients with psychiatric disorders, VolBrain and MRICloud can detect subtle structural differences in the brain.
Se han observado anomalías de volumen en las estructuras subcorticales, incluidos el hipocampo, la amígdala, el tálamo, el núcleo caudado, el putamen y el globo pálido, en la esquizofrenia (SZ) y el trastorno bipolar (BD); no todos los individuos con estos trastornos presentan tales cambios. Además, los patrones específicos y la gravedad de los cambios de volumen pueden variar entre individuos y en diferentes etapas de la enfermedad. El estudio tuvo como objetivo comparar los volúmenes de estas estructuras subcorticales entre sujetos sanos e individuos diagnosticados con SZ o BD. Se realizaron mediciones volumétricas del ventrículo lateral, globo pálido, núcleo caudado, putamen, hipocampo y amígdala mediante resonancia magnética en 52 sujetos sanos (HS), 33 pacientes con SZ y 46 pacientes con BD. Se utilizaron métodos de segmentación automática para analizar las imágenes de resonancia magnética con VolBrain y MRICloud. El hipocampo, la amígdala y el ventrículo lateral aumentaron en pacientes con esquizofrenia y trastorno bipolar en comparación con sujetos de control que utilizaron MRIcloud. El globo pálido y el núcleo caudado aumentaron en pacientes con esquizofrenia y trastorno bipolar en comparación con los sujetos control que utilizaron Volbrain. Sugerimos que en pacientes con esquizofrenia y trastorno bipolar, nuestros resultados contribuirán a la evaluación de la progresión subcortical, la patología y las anomalías de las composiciones cerebrales subcorticales. En pacientes con trastornos psiquiátricos, VolBrain y MRICloud pueden detectar diferencias estructurales sutiles en el cerebro.
Assuntos
Humanos , Masculino , Feminino , Adulto , Pessoa de Meia-Idade , Esquizofrenia/diagnóstico por imagem , Transtorno Bipolar/diagnóstico por imagem , Imageamento por Ressonância Magnética/métodos , Tamanho do Órgão , Esquizofrenia/patologia , Transtorno Bipolar/patologia , Estudos Transversais , Estudos Retrospectivos , Computação em NuvemRESUMO
This work presents Chameleon, a cloud computing (CC) Industry 4.0 (I4) neutron spectrum unfolding code. The code was designed under the Python programming language, using Streamlit framework®, and it is executed on the cloud, as I4 CC technology through internet, by using mobile devices with internet connectivity and a web navigator. In its first version, as a proof of concept, the SPUNIT algorithm was implemented. The main functionalities and the preliminary tests performed to validate the code are presented. Chameleon solves the neutron spectrum unfolding problem and it is easy, friendly and intuitive. It can be applied with success in various workplaces. More validation tests are in progress. Future implementations will include improving the graphical user interface, inserting other algorithms, such as GRAVEL, MAXED and neural networks, and implementing an algorithm to estimate uncertainties in the calculated integral quantities.
Assuntos
Algoritmos , Computação em Nuvem , Redes Neurais de Computação , Internet , NêutronsRESUMO
As startups são empresas que apresentam modelos de negócios marcados pela inovação, rapidez, flexibilidade e alta capacidade de adaptação aos mercados. Atuando em diferentes setores socioeconômicos, elas prometem criar e transformar produtos e serviços. A emergência e disseminação dessas empresas ocorrem em um momento histórico de mudanças iniciadas a partir de 1970 e marcadas pelas crises geradas com o esgotamento do paradigma da sociedade urbano industrial. No Brasil, o número desse modelo de negócio apresentou uma expansão expressiva, alcançando a marca de 13.374 nos últimos cinco anos. Atento a esse cenário, o objetivo desta pesquisa consistiu em compreender como sujeitos, grupos e instituições atribuem sentidos à experiência de trabalho nas chamadas startups. Na parte teórica, as condições sociais e econômicas que possibilitaram a emergência e disseminação das startups são analisadas em uma perspectiva crítica. A parte empírica, por sua vez, apresenta depoimentos de empreendedores relatando o contexto geral de atuação nas startups. Ao final deste artigo, conclui-se que há uma instrumentalização capitalística de componentes subjetivos específicos selecionados e colocados em circulação para fortalecer o modo de produção capitalista financeirizado.(AU)
Startups are companies that have business models characterized by innovation, speed, flexibility, and a high capacity to adapt to markets. Operating in different socioeconomic sectors, they promise to create and transform products and services. The emergence and dissemination of these companies occur at a historical moment of changes that began from 1970 and are marked by the crises generated by the exhaustion of the paradigm of industrial urban society. In Brazil, the number of businesses in this model showed a significant expansion, reaching 13,374 companies in the last five years. Attentive to this scenario, the objective of this research was to understand how subjects, groups, and institutions attribute meanings to the work experience in so-called startups. In the theoretical part, the social and economic conditions that enabled the emergence and dissemination of startups are analyzed in a critical perspective. The empirical part presents entrepreneurs reporting the general context of action in startups. At the end of this article, it is concluded that there is a capitalistic instrumentalization of specific subjective components that are selected and put into circulation to strengthen the financed capitalist production.(AU)
Las startups son empresas que tienen modelos de negocio marcados por la innovación, la velocidad, la flexibilidad y una alta capacidad de adaptación a los mercados. Desde diferentes sectores socioeconómicos, las startups prometen crear y transformar productos y servicios. La aparición y difusión de estas empresas se produce en un momento histórico de cambios que comenzó a partir de 1970 y que está marcado por crisis generadas por el agotamiento del paradigma de la sociedad urbana industrial. En Brasil, estas empresas se expandieron significativamente alcanzando la marca de 13.374 empresas en los últimos cinco años. En este escenario, el objetivo de esta investigación fue entender cómo los sujetos, grupos e instituciones atribuyen significados a la experiencia laboral en las startups. En la parte teórica, se analizan las condiciones sociales y económicas que permitieron el surgimiento y la difusión de las startups en una perspectiva crítica. La parte empírica presenta testimonios de emprendedores que informan sobre el trabajo en startups. La investigación concluye que hay una instrumentalización capitalista de componentes subjetivos específicos que se seleccionan y ponen en circulación para fortalecer el modo de producción capitalista financiero.(AU)
Assuntos
Humanos , Masculino , Feminino , Satisfação Pessoal , Psicologia Social , Trabalho , Organizações , Capitalismo , Organização e Administração , Inovação Organizacional , Grupo Associado , Personalidade , Política , Corporações Profissionais , Prática Profissional , Psicologia , Relações Públicas , Gestão de Riscos , Segurança , Salários e Benefícios , Ajustamento Social , Mudança Social , Valores Sociais , Tecnologia , Pensamento , Jornada de Trabalho , Tomada de Decisões Gerenciais , Proposta de Concorrência , Financiamento de Capital , Inteligência Artificial , Conferências de Consenso como Assunto , Cultura Organizacional , Saúde , Pessoal Administrativo , Saúde Ocupacional , Técnicas de Planejamento , Adolescente , Empreendedorismo , Readaptação ao Emprego , Setor Privado , Modelos Organizacionais , Entrevista , Gestão da Qualidade Total , Gerenciamento do Tempo , Eficiência Organizacional , Comportamento Competitivo , Recursos Naturais , Comportamento do Consumidor , Serviços Contratados , Benchmarking , Patente , Serviços Terceirizados , Evolução Cultural , Marketing , Difusão de Inovações , Competição Econômica , Eficiência , Emprego , Eventos Científicos e de Divulgação , Comercialização de Produtos , Estudos de Avaliação como Assunto , Agroindústria , Planejamento , Ensaios de Triagem em Larga Escala , Empresa de Pequeno Porte , Rede Social , Administração Financeira , Invenções , Crowdsourcing , Computação em Nuvem , Equilíbrio Trabalho-Vida , Participação dos Interessados , Crescimento Sustentável , Liberdade , Big Data , Utilização de Instalações e Serviços , Comércio Eletrônico , Blockchain , Desenho Universal , Realidade Aumentada , Inteligência , Investimentos em Saúde , Meios de Comunicação de Massa , OcupaçõesRESUMO
Recently, the number of vehicles equipped with wireless connections has increased considerably. The impact of that growth in areas such as telecommunications, infotainment, and automatic driving is enormous. More and more drivers want to be part of a vehicular network, despite the implications or risks that, for instance, the openness of wireless communications, its dynamic topology, and its considerable size may bring. Undoubtedly, this trend is because of the benefits the vehicular network can offer. Generally, a vehicular network has two modes of communication (V2I and V2V). The advantage of V2I over V2V is roadside units' high computational and transmission power, which assures the functioning of early warning and driving guidance services. This paper aims to discover the principal vulnerabilities and challenges in V2I communications, the tools and methods to mitigate those vulnerabilities, the evaluation metrics to measure the effectiveness of those tools and methods, and based on those metrics, the methods or tools that provide the best results. Researchers have identified the non-resistance to attacks, the regular updating and exposure of keys, and the high dependence on certification authorities as main vulnerabilities. Thus, the authors found schemes resistant to attacks, authentication schemes, privacy protection models, and intrusion detection and prevention systems. Of the solutions for providing security analyzed in this review, the authors determined that most of them use metrics such as computational cost and communication overhead to measure their performance. Additionally, they determined that the solutions that use emerging technologies such as fog/edge/cloud computing present better results than the rest. Finally, they established that the principal challenge in V2I communication is to protect and dispose of a safe and reliable communication channel to avoid adversaries taking control of the medium.
Assuntos
Segurança Computacional , Confidencialidade , Computação em Nuvem , Redes de Comunicação de Computadores , ComunicaçãoRESUMO
The rise of digitalization, sensory devices, cloud computing and internet of things (IoT) technologies enables the design of novel digital product lifecycle management (DPLM) applications for use cases such as manufacturing and delivery of digital products. The verification of the accomplishment/violations of agreements defined in digital contracts is a key task in digital business transactions. However, this verification represents a challenge when validating both the integrity of digital product content and the transactions performed during multiple stages of the DPLM. This paper presents a traceability method for DPLM based on the integration of online and offline verification mechanisms based on blockchain and fingerprinting, respectively. A blockchain lifecycle registration model is used for organizations to register the exchange of digital products in the cloud with partners and/or consumers throughout the DPLM stages as well as to verify the accomplishment of agreements at each DPLM stage. The fingerprinting scheme is used for offline verification of digital product integrity and to register the DPLM logs within digital products, which is useful in either dispute or violation of agreements scenarios. We built a DPLM service prototype based on this method, which was implemented as a cloud computing service. A case study based on the DPLM of audios was conducted to evaluate this prototype. The experimental evaluation revealed the ability of this method to be applied to DPLM in real scenarios in an efficient manner.
Assuntos
Blockchain , Internet das Coisas , Segurança Computacional , Computação em Nuvem , TecnologiaRESUMO
Cloud storage has become a keystone for organizations to manage large volumes of data produced by sensors at the edge as well as information produced by deep and machine learning applications. Nevertheless, the latency produced by geographic distributed systems deployed on any of the edge, the fog, or the cloud, leads to delays that are observed by end-users in the form of high response times. In this paper, we present an efficient scheme for the management and storage of Internet of Thing (IoT) data in edge-fog-cloud environments. In our proposal, entities called data containers are coupled, in a logical manner, with nano/microservices deployed on any of the edge, the fog, or the cloud. The data containers implement a hierarchical cache file system including storage levels such as in-memory, file system, and cloud services for transparently managing the input/output data operations produced by nano/microservices (e.g., a sensor hub collecting data from sensors at the edge or machine learning applications processing data at the edge). Data containers are interconnected through a secure and efficient content delivery network, which transparently and automatically performs the continuous delivery of data through the edge-fog-cloud. A prototype of our proposed scheme was implemented and evaluated in a case study based on the management of electrocardiogram sensor data. The obtained results reveal the suitability and efficiency of the proposed scheme.