Difference between revisions of "10-601B Perceptrons and Large Margin"

Latest revision as of 22:36, 8 February 2016

The difference between an on-line and batch algorithm.
The perceptron algorithm.
The importance of margins in machine learning.
The definitions of, and intuitions behind, these concepts:
- The margin of a classifier relative to a dataset.
- What a constrained optimization problem is.
- The SVM algorithm.
How the perceptron and SVM are similar and different.

@@ Line 3: / Line 3: @@
 === Slides ===
-*
+* [http://curtis.ml.cmu.edu/w/courses/images/8/82/Perceptron-svm_02_01.pptx Slides in powerpoint]
+* [http://curtis.ml.cmu.edu/w/courses/images/d/d2/Perceptron-svm_02_01.pdf Slides in pdf]
-=== Readings ===
+=== Useful Additional Readings ===
-* [http://www.cs.cmu.edu/~wcohen/10-601/vp-notes/vp.pdf My notes on the voted perceptron].  (You can skip sections 3-4 on ranking and the structured perceptron).
+* The Perceptron Algorithm: Mitchell 4.4.1 & 4.1.2, Bishop 4.1.7
-* Optional reading:  Freund, Yoav, and Robert E. Schapire. "Large margin classification using the perceptron algorithm." Machine learning 37.3 (1999): 277-296.
+* Support Vector Machines: Bishop 7.1, Murphy 14.5
-* Optional background on linear algebra: [http://www.cs.cmu.edu/~zkolter/course/linalg/ Zico Kolter's linear algebra review lectures]
 === What You Should Know Afterward ===
 * The difference between an on-line and batch algorithm.
-* How to implement the voted perceptron.
+* The perceptron algorithm.
-* The definition of a mistake bound, and a margin.
+* The importance of margins in machine learning.
+* The definitions of, and intuitions behind, these concepts:
+** The ''margin'' of a classifier relative to a dataset.
+** What a ''constrained optimization problem'' is.
+** The SVM algorithm.
+* How the perceptron and SVM are similar and different.