Navegando por Assunto "Machine learning"

Agora exibindo 1 - 3 de 3

Machine Learning e conservação da Amazônia: uma revisão sobre o uso de Machine Learning na conservação da região da Amazônia
(Centro Universitário do Estado do Pará, 2023-12-07) Dias, Carlos Eduardo Nylander Bitencourt; Barros, Rafael Luís Carvalho; Gomes, Vitor Hugo Freitas; Elgrably, Isaac Souza; http://lattes.cnpq.br/7590598824563858; http://lattes.cnpq.br/5218954387107307; Nascimento, Polyana Santos Fonseca; http://lattes.cnpq.br/6889523334917369; Araújo, Fábio Rocha de; http://lattes.cnpq.br/2407240421934932
With an approximated expanse of 6 million km², the Amazon Rainforest is a region of global interest, particularly for its great biodiversity some of which still unmeasured. The forest exercises an important part in regional and global climate control, capturing and storing CO2, contributing with rain formation and varied biogeochemical cycles. Its large contribution to the livelihood of native and traditional communities is not to be neglected. Despite it all, the territory faces great challenges, suffering with the impacts of deforestation, fires, global climate change, pollution, and more. In its conservation efforts, the usage of advancing tools and technology has resulted in the implementation of Artificial Intelligence and its sub-areas, such as Machine Learning, which has contributed to the development of predictive studies in the Amazon. Here, we revised studies that implemented Machine Learning in the conservation of Amazon’s territory, seeking a deeper understanding of its limitations and future usage of this technology. We concluded that, as a technology Machine Learning has helped in preservation efforts, but there still is much that can be done to improve its usage, such as the utilization of comprehensive data, training of professionals and experts to adequately implement the technology and to analyze the results, being these improvements crucial to the futureconservation efforts in the Amazon Region.
Otimização de hiperparâmetros de redes neurais por inferência Bayesiana com enxame de partículas
(Centro Universitário do Estado do Pará, 2023-12-07) Barros, Lucas Lima de Aragão; Souza, Daniel Leal; Mollinetti, Marco Antônio Florenzano; http://lattes.cnpq.br/1246642827046945; http://lattes.cnpq.br/6059334260016388; Nascimento, Polyana Santos Fonseca; http://lattes.cnpq.br/6889523334917369; Oliveira, Roberto Célio Limão de; http://lattes.cnpq.br/4497607460894318
This work explores - through hyperparameter optimization techniques - the possibility of improving the effectiveness of neural networks, which have experienced widespread popularity in various fields. The focus is to evaluate a new approach to hyperparameter optimization, which is one of the main challenges in the development of these networks. Traditionally, the definition of these hyperparameters occurs stochastically or through mathematical calculations. The goal is to refine the process, allowing a more precise definition of values, with lower losses. These solutions are based on a review of the literature on hyperparameter optimization in Neural Networks (e.g., CNN, Fully Connected), addressing fundamental concepts and techniques such as Grid Search, Random Search, and SMBO. The methodology includes the choice of tools, cross-validation strategy, and specific search approaches, such as Gaussian Regression or grid search. The detailed experiments involve optimizing the hyperparameters of Neural Networks, presenting datasets, model configurations, training protocols, and quantitative results. Despite its limitations and the need for further studies, this work demonstrates the feasibility and potential of combining advanced techniques for hyperparameter optimization in Neural Networks, offering contributions and stimulating new research directions in the search for greater efficacy in various applications of these networks. The discussion analyzes the effects of hyperparameter settings, identifies limitations, and addresses possible reasons for the obtained results. Finally, the conclusion summarizes the main insights, specific contributions, and future directions, while the references ensure the work's foundation.
Revisão sistemática sobre o uso de IA na produção de próteses de membros superiores
(Centro Universitário do Estado do Pará, 2022-06-08) Rosário, Jean Lucas Costa do; Assunção, Daniel Dias; Baganha, Alessandra Natasha Alcântara Barreiros; http://lattes.cnpq.br/0672275617505946; Nascimento, Polyana Santos Fonseca; http://lattes.cnpq.br/6889523334917369; Gomes, Vitor Hugo Freitas; http://lattes.cnpq.br/5218954387107307
Prostheses seek to replace the functions of the lost limb, so they are fundamental in the lives of people who have suffered from the loss of a limb, either congenitally or accidentally, especially when we talk about upper limb prostheses, due to the amount of tasks performed routinely. with hands and arms. Currently, there are several prostheses available, the myoelectric one being the most technologically advanced because it uses muscle signals for its activation, however this technology still has a deficit in the efficiency of performing the desired movement and has a high final cost. Considering that the efficiency of the prosthesis makes up this increase in the final cost along with the aesthetics and materials used, the present work analyzes the themes, starting from a systematic review, passing through the moment of the technology of upper limb prostheses, how artificial intelligence is contributing to the advancement of these technologies and what is its impact on the final cost of prostheses. It can be concluded that the advances in this topic are driven by the use of artificial intelligence, improving precision and control in the use of these upper limb prostheses, but the financial impact could not be verified.