Análisis de educaedu
Curso: Introduction to Statistical Learning (Introducción Al Aprendizaje Estadístico)
Comentarios sobre Curso: Introduction to Statistical Learning (Introducción Al Aprendizaje Estadístico) - Presencial - Almagro - Buenos Aires
Module 1: Statistical Learning I.
Learning. Inference. Machine learning: Generalization and Comprehensibility. Statistical
Learning. Supervised Learning. Unsupervised Learning. Components of a statistical learning
system: Statistical cost function, Optimization methods, Learning algorithm and Stopping
criterion. Limitations: Overfitting and Overtraining. Applications.
Module 2: Statistical Learning II.
Statistical modeling. Information Theory and Statistics. Algorithmic information theory.
Computational considetations. Inductive inference (Solomonoff theory). Occam's razor.. Ideal
MDL . Crude MDL. Refined MDL. Applications.
Module 3: Supervised learning: Linear regression models.
Least square estimation methods. Gauss-Markov theorem. Multiple regressions from univariate
regression. Subset selection methods. Ridge Regression. Lasso regression. Principal
Components Regression. Partial Least Squares. Comparison of the methods. Applications.
Module 4: Supervised learning: Linear methods for Classification.
Linear regression of an indicator matrix. Linear discriminant analysis (LDA. Regularized
discriminant analysis (RDA). Logistic regressio. Fitting logistic regression model. Separating
hyperplanes. Rosenblatt’s perceptron learning algorithm. Optimal separating hyperplanes.
Applications.
Module 5: Supervised learning: Kernel methods.
Kernel smoothers. Local Linear Regression. Local Polynomial Regression. Kernel Width.
Structured Local Regression. Local Likelihood. Kernel Density Estimation and Classification.
Radial Basis Functions. Applications.
Module 6: Supervised learning: Neural Networks.
Projection Pursuit Regression. Artificial Neural Networks. Multilayer Perceptron (MLP). Fitting
MLP. Issues in training Neural Networks. Radial Basis Function Networks (RBFN). Fitting
RBNF. Applications.
Module 7: Supervised learning: Support Vector Machines.
The Support Vector Classifier. SVM for classification. SVM and the Kernel property. SVM as a
penalization method. Reproducing Kernels. SVM for regression. Applications.
Module 8: Unsupervised learning: Association Rules.
Introduction. The Apriori algorithm. Unsupervised as supervised learning Generalized
Association Rules. Applications.
Module 9: Unsupervised learning: Cluster Análisis.
Dissimilarity measurement. Cluster algorithms. K-means. Vector quantization. K-medoids.
Hierarchical clustering. Applications.
Module 10: Unsupervised learning: Self-Organizing Maps.
SOM. Statistical considerations. Aplications.
Module 11: Principal component analysis.Principal components, curves and surfaces.
Aplications
Aclaración: El temario desarrollado del curso puede sufrir algunas pequeñas modificaciones, según lo crea conveniente el profesorado del mismo, con el fin de optimizar los tiempos y lograr la mejor comprensión posible de cada módulo por parte de los alumnos.
Opiniones (1)
Otra formación relacionada con investigación médica y de enfermedades
Centro: UBA - Facultad de Medicina
Solicita informaciónCurso - Talleres y Casos Clínicos sobre Diabetes
Centro: UBA - Facultad de Medicina
Solicita informaciónCurso - Intervenciones Asistidas con Perros
Centro: Asociación Argentina de Terapia Asistida con Perros
Solicita información
Un curso a mi parecer muy interesante.
Franco Silva
Curso: Introduction to Statistical Learning (Introducción Al Aprendizaje Estadístico) - Octubre 2011