# CS545 fall 2016

### Site Tools

assignments:assignment2

# Differences

This shows you the differences between two versions of the page.

 assignments:assignment2 [2016/09/06 09:54]asa assignments:assignment2 [2016/09/07 13:22]asa Both sides previous revision Previous revision 2016/09/14 09:38 asa [Assignment 2] 2016/09/07 13:22 asa 2016/09/06 09:54 asa 2016/09/06 09:51 asa 2016/09/06 09:40 asa 2016/09/04 10:19 asa [Grading] 2016/09/04 10:16 asa [Submission] 2016/09/04 10:13 asa 2016/09/04 10:12 asa 2016/09/04 09:55 asa 2016/08/30 09:27 asa 2016/08/30 09:22 asa 2016/08/30 09:21 asa 2016/08/30 09:20 asa 2016/08/30 09:19 asa 2016/08/27 11:50 asa 2016/08/27 11:27 asa 2016/08/27 11:24 asa 2016/08/27 11:19 asa 2016/08/27 11:18 asa 2016/08/27 11:17 asa 2016/08/09 10:25 external edit2015/09/28 12:24 asa 2015/09/25 13:52 asa 2015/09/16 16:52 asa 2015/09/16 16:50 asa 2015/09/16 16:47 asa 2015/09/16 16:45 asa 2016/09/14 09:38 asa [Assignment 2] 2016/09/07 13:22 asa 2016/09/06 09:54 asa 2016/09/06 09:51 asa 2016/09/06 09:40 asa 2016/09/04 10:19 asa [Grading] 2016/09/04 10:16 asa [Submission] 2016/09/04 10:13 asa 2016/09/04 10:12 asa 2016/09/04 09:55 asa 2016/08/30 09:27 asa 2016/08/30 09:22 asa 2016/08/30 09:21 asa 2016/08/30 09:20 asa 2016/08/30 09:19 asa 2016/08/27 11:50 asa 2016/08/27 11:27 asa 2016/08/27 11:24 asa 2016/08/27 11:19 asa 2016/08/27 11:18 asa 2016/08/27 11:17 asa 2016/08/09 10:25 external edit2015/09/28 12:24 asa 2015/09/25 13:52 asa 2015/09/16 16:52 asa 2015/09/16 16:50 asa 2015/09/16 16:47 asa 2015/09/16 16:45 asa 2015/09/16 09:47 asa 2015/09/16 09:38 asa 2013/09/20 14:11 asa 2013/09/20 14:04 asa 2013/09/20 13:43 asa 2013/09/20 13:37 asa 2013/09/20 13:13 asa 2013/09/20 12:39 asa 2013/09/20 12:35 asa 2013/09/19 12:57 asa 2013/09/19 12:54 asa created Last revision Both sides next revision Line 61: Line 61: The variable $\eta$ plays the role of the learning rate $\eta$ employed in the perceptron algorithm and $\delta \alpha$ is the proposed magnitude of change in $\alpha_i$. ​ The variable $\eta$ plays the role of the learning rate $\eta$ employed in the perceptron algorithm and $\delta \alpha$ is the proposed magnitude of change in $\alpha_i$. ​ We note that the adatron tries to maintain a //sparse// representation in terms of the training examples by keeping many $\alpha_i$ equal to zero.  The adatron converges to a special case of the SVM algorithm that we will learn later in the semester; this algorithm tries to maximize the margin with which each example is classified, which is captured by the variable $\gamma$ in the algorithm (notice that the magnitude of change proposed for each $\alpha_i$ becomes smaller as the margin increases towards 1). We note that the adatron tries to maintain a //sparse// representation in terms of the training examples by keeping many $\alpha_i$ equal to zero.  The adatron converges to a special case of the SVM algorithm that we will learn later in the semester; this algorithm tries to maximize the margin with which each example is classified, which is captured by the variable $\gamma$ in the algorithm (notice that the magnitude of change proposed for each $\alpha_i$ becomes smaller as the margin increases towards 1). + + **Note:** if you observe an overflow issues in running the adatron, add an upper bound on the value of $\alpha_i$. Here's what you need to do: Here's what you need to do:
assignments/assignment2.txt · Last modified: 2016/09/14 09:38 by asa