Difference between revisions of "10-601 Logistic Regression"

Latest revision as of 02:37, 18 February 2016

How to implement logistic regression.
How to determine the best parameters for logistic regression models
Why regularization matters for logistic regression.
How logistic regression and naive Bayes are similar and different.
The difference between a discriminative and a generative classifier.
What "overfitting" is, and why optimizing performance on a training set does not necessarily lead to good performance on a test set.

@@ Line 1: / Line 1: @@
-This a lecture used in the [[Syllabus for Machine Learning 10-601]]
+This a lecture used in the [[Syllabus for Machine Learning 10-601B in Spring 2016]]
 === Slides ===
+* William's lecture: [http://www.cs.cmu.edu/~wcohen/10-601/logreg.pptx in Powerpoint]  [http://www.cs.cmu.edu/~wcohen/10-601/logreg.pdf in PDF]
-* [http://www.cs.cmu.edu/~wcohen/10-601/logreg.pptx Slides in Powerpoint].  ''Based on the slides I used for 10-605, they might be updated.''
 === Readings ===
-* None
+* Optional:
+** Murphy 8.1-3, 8.6
-=== Assignment ===
+** [http://www.cs.cmu.edu/~wcohen/10-605/notes/sgd-notes.pdf William's notes on SGD] sec 1-3
+** [http://cseweb.ucsd.edu/~elkan/250B/logreg.pdf Charles Elkan's notes on SGD]
-* Implement logistic regression and apply it to a couple of datasets, using a gradient descent to optimize. Experiment by changing the regularization parameter. (Details to be posted later.)
 === What You Should Know Afterward ===
 * How to implement logistic regression.
-* Why regularization matters.
+* How to determine the best parameters for logistic regression models
+* Why regularization matters for logistic regression.
 * How logistic regression and naive Bayes are similar and different.
+* The difference between a discriminative and a generative classifier.
 * What "overfitting" is, and why optimizing performance on a training set does not ''necessarily'' lead to good performance on a test set.