It has become an annual tradition for Convolutional Neural Networks(CNNs)to continuously improve their performance in image classification and other applications.These advancements are often attributed to the adoption...It has become an annual tradition for Convolutional Neural Networks(CNNs)to continuously improve their performance in image classification and other applications.These advancements are often attributed to the adoption of more intricate network architectures,such as modules and skip connections,as well as the practice of stacking additional layers to create increasingly complex networks.However,the quest to identify the most optimizedmodel is a daunting task,given that stateof the artConvolutionalNeuralNetwork(CNN)models aremanually engineered.In this research paper,we leveraged a conventional Genetic Algorithm(GA)to craft optimized Convolutional Neural Network(CNN)architectures and pinpoint the ideal set of hyper parameters for image classification tasks using the MNIST dataset.Our experimentation with the MNIST dataset yielded remarkable results.Compared to earlier semi-automatic and automated approaches,our proposed GA demonstrated its efficiency by swiftly identifying the perfect CNN design,accomplishing this feat in just 6 GPU days while achieving an outstanding accuracy of 95.50%.展开更多
One Pr（Ⅲ） lanthanide ion complex was initially synthesized and characterized by TGA-DSC in air atmosphere, as well as characterized by CHN elemental analysis, defining the stoichiometric ratio as Pr（DMBz）3. The g...One Pr（Ⅲ） lanthanide ion complex was initially synthesized and characterized by TGA-DSC in air atmosphere, as well as characterized by CHN elemental analysis, defining the stoichiometric ratio as Pr（DMBz）3. The gaseous products evolved during the thermal decomposition were also monitored in N2 atmosphere employing TGA/FT-IR system. A crystal structure is obtained by state-of-the-art powder X-rays diffraction methods measured in conventional laboratory equipment and refined by the Rietveld method, which defined it as a monoclinic system of the space group P21/C with a polymeric crystal structure, [Pr（DMBZ）3]n. FT-IR theoretical spectrum and time-dependent density functional theory（TD-DFT） were calculated from TGA-DSC and crystalline system data. The experimental and theoretical FT-IR spectra present a high correlation degree when the main stretching bands are compared, while the energy transfer（HOMO - LUMO） in their neighborhoods suggests the main contributions of the light-emitting states.展开更多
文摘It has become an annual tradition for Convolutional Neural Networks(CNNs)to continuously improve their performance in image classification and other applications.These advancements are often attributed to the adoption of more intricate network architectures,such as modules and skip connections,as well as the practice of stacking additional layers to create increasingly complex networks.However,the quest to identify the most optimizedmodel is a daunting task,given that stateof the artConvolutionalNeuralNetwork(CNN)models aremanually engineered.In this research paper,we leveraged a conventional Genetic Algorithm(GA)to craft optimized Convolutional Neural Network(CNN)architectures and pinpoint the ideal set of hyper parameters for image classification tasks using the MNIST dataset.Our experimentation with the MNIST dataset yielded remarkable results.Compared to earlier semi-automatic and automated approaches,our proposed GA demonstrated its efficiency by swiftly identifying the perfect CNN design,accomplishing this feat in just 6 GPU days while achieving an outstanding accuracy of 95.50%.
基金Project supported by Coordination for the Improvement of Higher Education Personnel(CAPES)the Brazilian Foundations CNPQ(474259/2013-7)+1 种基金FUNDECT(23/200.632/2014)Federal University of Grande Dourados
文摘One Pr（Ⅲ） lanthanide ion complex was initially synthesized and characterized by TGA-DSC in air atmosphere, as well as characterized by CHN elemental analysis, defining the stoichiometric ratio as Pr（DMBz）3. The gaseous products evolved during the thermal decomposition were also monitored in N2 atmosphere employing TGA/FT-IR system. A crystal structure is obtained by state-of-the-art powder X-rays diffraction methods measured in conventional laboratory equipment and refined by the Rietveld method, which defined it as a monoclinic system of the space group P21/C with a polymeric crystal structure, [Pr（DMBZ）3]n. FT-IR theoretical spectrum and time-dependent density functional theory（TD-DFT） were calculated from TGA-DSC and crystalline system data. The experimental and theoretical FT-IR spectra present a high correlation degree when the main stretching bands are compared, while the energy transfer（HOMO - LUMO） in their neighborhoods suggests the main contributions of the light-emitting states.
