Difference between revisions of "10-601 Logistic Regression"
From Cohen Courses
Jump to navigationJump to search| Line 12: | Line 12: | ||
=== Assignment === | === Assignment === | ||
| − | * Implement logistic regression and apply it to a couple of datasets, using a gradient descent to optimize. Experiment by changing the regularization parameter. | + | * Implement logistic regression and apply it to a couple of datasets, using a gradient descent to optimize. Experiment by changing the regularization parameter and learning rate. |
=== What You Should Know Afterward === | === What You Should Know Afterward === | ||
Revision as of 09:49, 18 September 2013
This a lecture used in the Syllabus for Machine Learning 10-601
Slides
Readings
- William's notes on SGD (for 10605)
- Lazy Sparse Stochastic Gradient Descent for Regularized Multinomial Logistic Regression, Carpenter, Bob. 2008. See also his blog post on logistic regression.
Assignment
- Implement logistic regression and apply it to a couple of datasets, using a gradient descent to optimize. Experiment by changing the regularization parameter and learning rate.
What You Should Know Afterward
- How to implement logistic regression.
- Why regularization matters.
- How logistic regression and naive Bayes are similar and different.
- What "overfitting" is, and why optimizing performance on a training set does not necessarily lead to good performance on a test set.
