Difference between revisions of "10-601 Logistic Regression"
From Cohen Courses
Jump to navigationJump to search| Line 16: | Line 16: | ||
* Why regularization matters. | * Why regularization matters. | ||
* How logistic regression and naive Bayes are similar and different. | * 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. | * What "overfitting" is, and why optimizing performance on a training set does not ''necessarily'' lead to good performance on a test set. | ||
Revision as of 12:48, 24 September 2013
This a lecture used in the Syllabus for Machine Learning 10-601
Slides
Readings
- William's notes on SGD (for 10605)
- Charles Elkan's notes on SGD
- Lazy Sparse Stochastic Gradient Descent for Regularized Multinomial Logistic Regression, Carpenter, Bob. 2008. See also his blog post on logistic regression.
What You Should Know Afterward
- How to implement logistic regression.
- Why regularization matters.
- 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.
