The Laboratory of Computational Intelligence and Statistical Learning (LICAE) aims to study, propose and evaluate techniques based on Computational Intelligence and Statistical Learning to solve problems in Agriculture. Such proposals aim to improve existing methodologies which are applied routinely. To achieve this objective, LICAE aims at the formation of study groups, extension courses and preparation of materials such as the publication of articles and books in order to present theoretical aspects and practical applications in Agriculture in relation to themes that include Computational Intelligence and Statistical Learning .
The use of this set of techniques in agriculture is still recent, but there is already evidence that these techniques can be used in order to replace or complement traditional biometric techniques.
The LICAE team is multidisciplinary, working in Applied Statistics and Biometrics, Genetics and Breeding, Applied Economics and Animal Science.
Projects under development include: Use of Artificial Neural Networks in Broad Genomic Selection and Analysis of Adaptability and Phenotypic Stability; Regression, Bagging, Boosting and Random Forest trees to predict and indicate better crossings; Quantile Regression Methods in Association Analysis and Genomic Selection; Regularization methods, among others.