Machine Learning 202
We will cover several machine learning topics in detail. Among those are Bayesian Belief Networks, Prof. Jerome Friedman's gradient boosting papers, svd's and recommender systems.
For Machine Learning 202, we assume you are familiar with basic statistical concepts and have the ability to write programs running different algorithms on public data sets. We assume knowledge at the level of "Introduction to Data Mining" by Tan. If you have taken our Machine Learning
102 classes and Machine Learning
201, you are well prepared for this course.
We expect you to have previously used R. We will use R for discussing homework problems and comparing different solution approaches. .
http://cran.r-project.org/ For your review, R are here:
References for R,
Reference for R Comments,
More R references. To integrate R with Eclipse
The class will meet:
Wednesday and Thursday evenings.
from 7:00 pm to 9:00 pm
from 8/24/2011 - 9/22/2011
For a total of 10 meetings
Machine Learning 202 - Course Outline
Week 1: Collaborative Filtering, Recommendation Engines
Week 2: Bayesian Belief Networks, EM & Factor Analysis
Week 3: Advanced Trees
Week 4: Gradient Boosting
Week 5: Learning Theory, Debugging Methods
The book for the course is:
Elements of Statistical Learning by Hastie, Tibshirani and Friedman
The cost for full participation in the 10 class meetings is $300 or $200 for hacker dojo members. Hacker dojo members may participate in evening classes for free.