62 documents found, page 1 of 7

SecureQNN: Shielding the intellectual property of QNNs in TinyML systems

Building accurate Machine Learning (ML) models requires substantial expertise and large-scale datasets typically only available in big data companies. These companies have been selling their models as Machine Learning as a Service. However, concerns about data privacy and the appliance of ML in mission-critical scenarios are forcing ML computation to move from the cloud to the deep edge, near sensor data. If ed...

Mértola (Portugal) entre Roma y el Islam. Evidencias de las transformaciones de...

The excavations of recent years in Mértola, Portugal, enabled the discovery of new contexts of occupation of the city, that attest the importance and dynamism of the city in Late Antiquity, also the continuities of its structures in the first centuries of Islamic rule and its slow transformations, and the new elements that were introduced into the urban landscape, especially from the 9th century. In this paper ...

Um banho privado na Mértola Islâmica: uma interpretação preliminar no contexto ...

Durante as escavações arqueológicas realizadas na Encosta do Castelo de Mértola, no mês de julho de 2023, foram encontradas estruturas correspondentes a uns pequenos banhos de época almóada (primeiras décadas do século XIII). Embora a escavação ainda não tenha sido concluída, numa parte significativa da sua área, foi possível identificar, sem lugar a dúvidas, o compartimento destinado à sala quente dos banhos e...

Securing AI with AI: Novel framework for drone communication security

Autonomous drone swarms operating in potentially hostile environments and communicating over inherently insecure wireless channels require robust security architectures. While AI-based algorithms are effective in detecting communication anomalies and intrusions, their deployment in low-power environments like drones is challenging, and the integrity of AI-driven decisions can also be compromised. This paper dis...

Protecting the intellectual property of QNNs at the Deep Edge

This study presents the first framework designed to protect the intellectual property of Quantized Neural Networks (QNNs) on Arm Cortex-M MCUs equipped with TrustZone-M. Our framework- Secure QNN- employs an iterative simulation to identify the most privacy critical layers of a QNN, strategically offloading their execution to the secure-world of TrustZone-M. For each set of private layers, SecureQNN computes th...

Childhood neglect: Another perspective on childhood obesity

Childhood obesity is a multifactorial condition. Extreme cases are often associated with the inability of caregivers to follow the recommended diet plan, despite prior warnings of the potential risks associated of non-compliance. A seven-year-old girl with a body mass index (BMI) of 38.6 kg/m2 (z-score +7.3) and multiple comorbidities was seen in a Pediatric Nutrition outpatient consultation. After several atte...

O papel da IIoT na otimização de processos produtivos

O papel da IIoT na otimização de processos produtivos

Childhood neglect: Another perspective on childhood obesity

Childhood obesity is a multifactorial condition. Extreme cases are often associated with the inability of caregivers to follow the recommended diet plan, despite prior warnings of the potential risks associated of non-compliance.A seven-year-old girl with a body mass index (BMI) of 38.6 kg/m2 (z-score +7.3) and multiple comorbidities was seen in a Pediatric Nutrition outpatient consultation. After several attem...

Stromal Ascorbate Peroxidase (OsAPX7) Modulates Drought Stress Tolerance in Ric...

Chloroplast ascorbate peroxidases exert an important role in the maintenance of hydrogen peroxide levels in chloroplasts by using ascorbate as the specific electron donor. In this work, we performed a functional study of the stromal APX in rice (OsAPX7) and demonstrated that silencing of OsAPX7 did not impact plant growth, redox state, or photosynthesis parameters. Nevertheless, when subjected to drought stress...