10/09/2013
Stat841 - Wiki Course Notes
wikicoursenote.com/w/index.php?title=Stat841&printable=yes
1/74
Stat841
From Wiki Course Notes
Contents
1 Proposal
2 Mark your contribution here
3 Scribe sign up
4 Classfication-2009.9.30
4.1 Classification
4.2 Error rate
4.3 Bayes Classifier
4.4 Bayesian vs. Frequentist
5 Linear and Quadratic Discriminant Analysis - October 2,2009
5.1 Introduction
5.1.1 Notation
5.1.2 Approaches
5.2 LDA
5.2.1 Motivation
5.2.2 History
5.2.3 Definition
5.2.4 Limitation
5.3 QDA
6 Linear and Quadratic Discriminant Analysis cont'd - October 5, 2009
6.1 Summarizing LDA and QDA
6.2 In practice
6.3 Computation
6.4 The Number of Parameters in LDA and QDA
7 LDA and QDA in Matlab - October 7, 2009
8 Trick: Using LDA to do QDA - October 7, 2009
8.1 Motivation
8.2 Theoretically
8.3 By Example
9 Introduction to Fisher's Discriminant Analysis - October 7, 2009
9.1 Contrasting FDA with PCA
9.2 Intuitive Description of FDA
9.3 Example in R
9.4 Distance Metric Learning VS FDA
10 Fisher's Discriminant Analysis (FDA) - October 9, 2009
10.1 Two-class problem
10.1.1 Between class covariance
10.1.2 Within class covariance
10.1.3 Objective Function
10.1.4 FDA vs. PCA Example in Matlab
10.1.5 Practical example of 2_3
10.1.6 An extension of Fisher's discriminant analysis for stochastic processes
11 FDA for Multi-class Problems - October 14, 2009
11.1 FDA method for Multi-class Problems
11.2 Generalization of Fisher's Linear Discriminant
12 Linear Regression Models - October 14, 2009
12.1 A linear regression example in Matlab
12.2 Comments about Linear regression model
13 Logistic Regression- October 16, 2009
13.1 logistic function
13.2 Intuition behind Logistic Regression
13.3 The Logistic Regression Model
13.4 Fitting a Logistic Regression
13.4.1 Advantages and Disadvantages
13.4.2 Extension

This preview has intentionally blurred sections.
Sign up to view the full version.
10/09/2013
Stat841 - Wiki Course Notes
wikicoursenote.com/w/index.php?title=Stat841&printable=yes
2/74
14 Logistic Regression(2) - October 19, 2009
14.1 Logistic Regression Model
14.2 Find
14.2.1 WLS
14.2.2 Pseudo Code
14.3 Comparison with Linear Regression
14.4 Comparison with LDA
14.4.1 By example
15 2009.10.21
15.1 Multi-Class Logistic Regression
15.2 Multi-class kernel logistic regression
15.3 Perceptron (Foundation of Neural Network)
15.3.1 Separating Hyperplane Classifiers
15.3.2 Perceptron
15.3.3 A Perceptron Example
16 The Perceptron (Lecture October 23, 2009)
16.1 Problems with the Algorithm and Issues Affecting Convergence
16.2 Comment on gradient descent algorithm
17 Neural Networks (NN) - October 28, 2009
17.1 Introduction
17.2 Activation Function
17.3 Back-propagation
18 Neural Networks (NN) - October 30, 2009
18.1 Back-propagation
18.1.1 How to initialize the weights
18.1.2 How to set learning rates
18.1.3 How to determine the number of hidden units
18.2 Dimensionality reduction application
18.3 Deep Neural Network
18.3.1 Difficulties of training deep architecture
[12]
18.4 Neural Networks in Practice
18.5 Issues with Neural Network
18.6 BUSINESS APPLICATIONS OF NEURAL NETWORKS
19 Complexity Control October 30, 2009
20 Complexity Control - Nov 2, 2009
20.1 How do we choose a good classifier?


This is the end of the preview.
Sign up
to
access the rest of the document.
- Winter '13
- Conditional Probability, Quadratic classifier
-
Click to edit the document details