Introduction to Machine Learning

Transcription

1 Introduction to Machine Learning CS 590 and STAT 598A, Spring 2010 Instructor: S.V. N. Vishwanathan ( vishy) January 12, 2010 S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 1 / 17

2 Class Details Classes: Tue/Thurs 9:00 am - 10:15 am LWSN B134 S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 2 / 17

3 Textbook Introduction to Machine Learning Alex Smola and S.V.N. Vishwanathan Yahoo Labs Santa Clara and Departments of Statistics and Computer Science Purdue University and College of Engineering and Computer Science Australian National University S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 3 / 17

4 Supplementary Textbook S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 4 / 17

5 Course Description This is an introductory course in machine learning You will learn about a number of basic machine learning algorithms such as k-means k-nearest neighbors Perceptron naive Bayes EM You will also some fairly modern topics such as Support Vector Machines Gaussian Processes Exponential Families Conditional Random Fields Graphical Models Structured Prediction Emphasis throughout the course will be on connections between various algorithms S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 5 / 17

6 Ideal Audience Well versed with fundamental statistical concepts such as Probability Random Variables Mean and Variance etc. Comfortable with statistical algorithms such as Linear and Logistic regression k-means clustering etc. Good familiarity with a high level programming language such as C, C++, or Python. In a pinch Matlab or R will do (but not recommended). Interested in learning how to efficiently code algorithms for large scale data analysis S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 6 / 17

7 Prerequisites Required Basic Probability and Applications MA 511: Linear Algebra Programming in some high level language Or equivalent... S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 7 / 17

8 Grading Policy 5 Assignments: 10 points each Course project: 25 points Midterm: 20 points Class Participation: 5 points Other policies on the course home page. Please review carefully. S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 8 / 17

9 Office Hours Office Hours: 2:00-3:00pm Tue or by appointment at HAAS 232 S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 9 / 17

10 Frequently Asked Questions I Q: Will I need to do lots of programming? Ans: Yes. This is a very hands on course and will involve coding different machine learning algorithms. You should budget significant amounts of time for your assignments. S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 10 / 17

11 Frequently Asked Questions II Q: Will I need lots of maths to understand your lectures? Ans: I expect familiarity with Linear Algebra Multivariate Calculus Probability Theory as pre-requisites. There will be emphasis on rigor even when learning about machine learning algorithms. Q: Can I meet you anytime I want? Ans: I will definitely be around during office hours. You are welcome to walk in any other time I am in my office, but do remember that I generally have busy days. To avoid disappointment it is best to book a slot via . S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 11 / 17

12 Frequently Asked Questions III Q: Do you reply to s? Ans: I try to reply to s as promptly as possible. If you do not hear back from me within 3-4 days then please ping me during the class. Your may have ended up in my junk mail folder! Q: Can I solve the HW problems collaboratively? Ans: The course policy clearly says: Group discussions are encouraged to further understand difficult topics. You may consult with other students about homework problems, provided that you indicate such information (whom you consulted with, which problem, to which extent) on your solution sheet. However, you must refrain from getting direct answers from others. Any violation will result in zero credit for the assignment. S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 12 / 17

13 Frequently Asked Questions IV Q: How do I submit my HW? Ans: For problems which do not involve coding, neatly type or write the solution and submit in class. I strongly encourage the use of LaTeX and discourage the use of MS Word. For solutions which involve coding, submit a print out in class and send your code via before the class. Q: How will you evaluate the project? Ans: First you need to choose a project topic and discuss it with me. Then you make a proposal which lays out the what you will deliver. After the project you will need to submit your code and give a short presentation. You will be evaluated against: What you promised and how much you delivered. Magnitude of your effort vis-a-vis the rest of the class Your presentation S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 13 / 17

14 Frequently Asked Questions V Q: Will you post notes for all topics? Ans: Yes for almost all topics except standard ones for which I will refer you to chapters in a text book or to other standard resources. Q: Will you use slides (e.g. powerpoint) for your lectures? Ans: No. I prefer to lecture on the blackboard. Class notes will be available for download from the course home page shortly after the class. S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 14 / 17

15 Topics (Tentative) Review Density Estimation Exponential families of distributions Directed and Undirected Graphical models Structured Learning Optimization for Machine Learning Kernels S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 15 / 17

16 Background Survey Please answer as truthfully as possible Can help me tailor the lectures Talk to me if you have any concerns or comments S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 16 / 17

17 Thank You! Questions? S.V. N. Vishwanathan (Purdue University) Introduction to Machine Learning 17 / 17