@InProceedings{CarrubaAljCarDomMar:2022:ClAsRe,
               author = "Carruba, Val{\'e}rio and Aljbaae, Safwan and Carit{\'a}, Gabriel 
                         Antonio and Domingos, R. C. and Martins, B.",
          affiliation = "{Universidade Estadual Paulista (UNESP)} and {Instituto Nacional 
                         de Pesquisas Espaciais (INPE)} and {Instituto Nacional de 
                         Pesquisas Espaciais (INPE)} and {Universidade Estadual Paulista 
                         (UNESP)} and {Universidade Estadual Paulista (UNESP)}",
                title = "Classification of asteroids’ resonant arguments using 
                         Convolutional Neural Networks",
                 year = "2022",
         organization = "Col{\'o}quio Brasileiro de Din{\^a}mica Orbital, 221.",
             abstract = "The asteroidal main belt is crossed by a web of mean-motion and 
                         secular resonances, that occur when there is a commensurability 
                         between fundamental frequencies of the asteroids and planets. 
                         Traditionally, these objects were identified by visual inspection 
                         of the time evolution of their resonant argument, which is a 
                         combination of orbital elements of the asteroid and the perturbing 
                         planet(s). Since the population of asteroids affected by these 
                         resonances is, in some cases, of the order of several thousand, 
                         this has become a taxing task for a human observer. Recent works 
                         used Convolutional Neural Networks (CNN) models to perform these 
                         tasks automatically. Here, we compare the outcome of such models 
                         with those of some of the most advanced and publicly available CNN 
                         architectures, like the VGG, Inception and ResNet. The performance 
                         of these models is first tested and optimized for overfitting 
                         issues, using validation sets and a series of regularization 
                         techniques like data augmentation, dropout, and batch 
                         normalization. The three best-performing models were then used to 
                         predict the labels of larger testing databases containing 
                         thousands of images. The VGG model, with and without 
                         regularizations, proved to be the most efficient method to predict 
                         labels of large datasets. Applications of such methods to 
                         asteroids interacting with secular and mean-motion resonances, 
                         like the \ν6 and M1:2 exterior resonance with Mars, already 
                         produced significant discoveries, like the identification of the 
                         (12988) Tiffanykapler asteroid family. This is the first young 
                         asteroid family ever found in a linear secular resonance, for 
                         which precise estimates of both the age and the ejection velocity 
                         field can be obtained. Since the Vera C. Rubin observatory is 
                         likely to discover up to four million new asteroids in the next 
                         few years, the use of CNN models might become quite valuable to 
                         identify populations of resonant minor bodies.",
  conference-location = "12-16 dez. 2022",
      conference-year = "S{\~a}o Jos{\'e} dos Campos, SP",
             language = "pt",
        urlaccessdate = "05 jun. 2024"
}