• Login
    View Item 
    •   DSpace Home
    • Trabalhos de Especialização
    • Ciências Exatas e da Terra
    • Inteligência Artificial Aplicada
    • View Item
    •   DSpace Home
    • Trabalhos de Especialização
    • Ciências Exatas e da Terra
    • Inteligência Artificial Aplicada
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Memorial de projetos : aprendizado profundo: uma visão geral sobre arquiteturas neurais modernas

    Thumbnail
    View/Open
    R - E - LUIS HENRIQUE PAIVA.pdf (2.927Mb)
    Date
    2025
    Author
    Paiva, Luís Henrique
    Metadata
    Show full item record
    Abstract
    Resumo: As redes neurais formam um dos pilares do Aprendizado Profundo. Esses algoritmos funcionam de forma similar ao cérebro humano e possibilitam que problemas reais possam ser resolvidos, como reconhecimento de imagens processamento de linguagem natural, sistemas de recomendação, robótica, reconhecimento de padrões em texto e áudio, entre outros. Este trabalho está dividido em duas partes. Na primeira, o parecer técnico, é apresentada uma visão geral sobre as arquiteturas RNN (Rede Neural Recorrente), CNN (Rede Neural Convolucional), GAN (Rede Adversária Generativa) e Transformer, passando pela sua definição, principais características e principais caso de aplicação. O parecer conclui retomando os principais conceitos discutidos nele e finaliza indicando a importância e o impacto dessas estruturas na melhoria de processos e atividades humanas, além dos novos desafios que se apresentam quando essas arquiteturas são expostas a novos tipos de problemas. Na segunda, e última parte, são apresentados os apêndices com as principais resoluções desenvolvidas durante cada disciplina do curso, além dos resultados aos questionamentos propostos
     
    Abstract: Neural networks are one of the pillars of Deep Learning. These algorithms work in a similar way to the human brain and enable real-world problems to be solved, such as image recognition, natural language processing, recommendation systems, robotics, pattern recognition in text and audio, among others. This paper is divided into two parts. The first, the technical review, presents an overview of the RNN (Recurrent Neural Network), CNN (Convolutional Neural Network), GAN (Generative Adversarial Network), and Transformer architectures, covering their definitions, main characteristics, and key application cases. The review concludes by reviewing the main concepts contained therein and concluding by highlighting the importance and impact of these structures in improving human processes and activities, as well as the new challenges that arise when these architectures are exposed to new types of problems. In the second and final part, appendices are presented with the main resolutions developed during each course discipline, in addition to the results of the proposed questions
     
    URI
    https://hdl.handle.net/1884/98765
    Collections
    • Inteligência Artificial Aplicada [143]

    DSpace software copyright © 2002-2022  LYRASIS
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Browse

    All of DSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsxmlui.ArtifactBrowser.Navigation.browse_typeThis CollectionBy Issue DateAuthorsTitlesSubjectsxmlui.ArtifactBrowser.Navigation.browse_type

    My Account

    LoginRegister

    Statistics

    View Usage Statistics

    DSpace software copyright © 2002-2022  LYRASIS
    Contact Us | Send Feedback
    Theme by 
    Atmire NV