/MO444-PatternRecognition-and-MachineLearning

MO444 (2nd Semester, 2015) - Pattern Recognition and Machine Learning

Primary LanguagePythonMIT LicenseMIT

MO444-PatternRecognition-and-MachineLearning

MO444 (2nd Semester, 2015) - Pattern Recognition and Machine Learning

Disciplina ministrada pelo Professor Dr. Andersonde Rezende Rocha

Para acesso de material da disciplina: http://www.ic.unicamp.br/~rocha/teaching/2015s2/mo444/index.html

Modelo de artigo: report-model.zip

4 trabalhos + Trabalho Final

Todos os trabalhos devem ser preparados os artigos científicos para entrega.

  1. 2015s2-mo444-assignment-01
  2. 2015s2-mo444-assignment-02
  3. 2015s2-mo444-assignment-03
  4. 2015s2-mo444-assignment-04
  5. Trabalho Final

Os assuntos abordados são:

Class Material #00

  1. Presentation of the discipline. Syllabus.
  2. Introduction Class -- Introduction to Machine Learning, problems, data, tools.
  3. Reading: IAAM, Chapter #1 e #2; PRML, Chapter #1

Class Material #01

  1. Introduction to ML
  2. Supervised Learning vs Unsupervised Learning vs Semi-Supervised Learning
  3. Liner Regression
  4. Cost Function
  5. Gradient Descent
  6. Generalization of Gradient Descent
  7. Model Complexity
  8. Overfitting vs. Generalization
  9. Multi-variate Regression
  10. Normalization
  11. Polynomial Regression
  12. Normal Equations vs. Gradient Descent
  13. Logistic Regression
  14. Decision Boundaries
  15. Logistic Regression and Cost Function
  16. Logistic Regression and Multi-class extensions
  17. Regularization
  18. Regularized Linear Regression and Logistic Regression

Class Material #02

  1. Perceptron
  2. Effects of Dimensionality
  3. Neural Networks
  4. Cost Function
  5. Backpropagation
  6. Gradient Checking

Class Material #03

  1. Unsupervised Learning
  2. Clustering
  3. K-Means
  4. Hard vs. Soft Assignment
  5. Gaussian Mixture Models (GMMs)
  6. Expectation/Maximization (EM)
  7. Dimensionality Reduction
  8. PCA and LDA
  9. Multi-class LDA

Class Material #04

  1. Evolutionary Computing
  2. Genetic Algorithms
  3. Genetic Programming
  4. Evolutionary Programming
  5. Evolutionary Strategies
  6. Operators
  7. Problem Examples

Class Material #05

  1. Data Representation vs. Data Classification
  2. Debugging an ML solution
  3. Performance Evaluation
  4. Bias vs. Variance
  5. ROC curves
  6. Bootstrapping
  7. Statistical Tests
  8. Wilcoxon Sign-Rank Test
  9. Friedman Test
  10. Post-tests

Class Material #06

  1. Decision tree learning

Class Material #07

  1. Sampling Theory
  2. Bagging
  3. Boosting

Class Material #08 #09 #10

  1. Support Vector Machines (I)
  2. Support Vector Machines (II)
  3. Support Vector Machines (III)

Class Material #11

  1. Random Forests (I)
  2. Random Forests (II)

Class Material #13

  1. Naive Bayes

Class Material #15

  1. Deep Learning

Class Material #16

  1. Optimum-Path Forest Classifier (OPF)