Course»Course 6»Fall 2012»6.S080»Homepage

6.S080  Introduction to Inference

Fall 2012

home page image

Instructors: Gregory W Wornell, Konstantinos Daskalakis, Polina Golland, Lizhong Zheng

TAs: George H Chen, Gauri Joshi, Ramesh Sridharan

Lecture:  MW10  (56-154)
Recitation:  TR1 or TR3  (34-303)
Office Hours (TA):  F3:30-4:30, M4:30-5:30  (24-322)
Office Hour (Greg):  W1:30-2:30  (36-677)
Office Hour (Lizhong):  R2-3  (36-660)

Information: 

Introduction to probabilistic modeling for problems of inference from data, emphasizing analytical and computational aspects. Distributions, marginalization, conditioning, and structure; graphical representations. Belief propagation, decision-making, classification, estimation, and prediction. Sampling methods and analysis. Introduction to asymptotic analysis and information measures, and applications.

(6-2) Foundation subject credit in either EE or CS

Enrollment limited

Announcements

Final exam prep info

Hi all,

To help you with your final exam preparation, the TAs will be providing the following:
  • Optional review session: Friday, 1-2:30 @ 34-301 (Gauri & Ramesh)
  • Office hours:
    • Friday, 3:30-4:30 @ 24-322 (Gauri)
    • Saturday, 3-5pm @ 32-D4 (George)
    • Sunday, 4-6 pm @ 32-D4 (Ramesh)
These are all optional, but feel free to come to as many as you wish.

The review session will be informal and focused on going over practice problems (which will likely be posted by Thursday night). Solutions and notes will be posted after the review session.

Ramesh

P.S. To get to the Saturday & Sunday office hours, take the Dreyfoos elevators in Stata to the 4th floor, where you'll see us through the glass. Knock and we'll let you in.

P.P.S. The review session and office hours will be your last chance to pick up graded psets.

Announced on 12 December 2012  4:22  p.m. by Ramesh Sridharan

Problem Set 9 - Due on 4th Dec

Hi all,

Problem set 9 is now posted on Stellar. Please note that it is due after two weeks, on 4th Dec. The next Pset will be posted on 4th Dec, but you are not required to turn it in.

Best,
Gauri.

Announced on 20 November 2012  2:24  p.m. by Gauri Joshi

Quiz 2 is on Wednesday November 7

Our second quiz is this coming Wednesday, Nov. 7, 7:30-9:30, in 32-141. Note that this is NOT our regular lecture room!

As summarized in the course information sheet, the quiz is closed book, but you may bring *two* 8.5x11" sheets of notes (both sides) with you to use. As before, calculators *will* be allowed but will probably not be helpful.

The quiz will cover material up to and including yesterday's lecture 15 (Wed., Oct. 31), and its recitation today (Thurs., Nov. 1), and the homework through Problem Set 7. It will focus on the material since the first quiz.

Also, the Tuesday (Nov. 6) recitation will be a pre-quiz review.

Please contact any of the staff if you have questions!

Announced on 01 November 2012  9:04  a.m. by Gregory W Wornell

Problem Set 7 has a mandatory anonymous survey to go with it

Hello all.

Problem Set 7, which was released earlier today, has been updated to have a mandatory anonymous survey:

1. It's mandatory in the sense that you need to turn it in to have your Problem Set 7 graded.

2. It's anonymous in that only the grader will know who turned in surveys (so please have your survey response attached to your PS7 solutions) but the grader will tear off the survey sheets from your submitted PS7's so that the TA's (who will read over the survey results) will not know who submitted which survey.

Detailed feedback is valuable to us--after all this is a pilot class!

Thanks,
George

Announced on 30 October 2012  3:03  p.m. by George H Chen

Recitation 11 notes on naive Bayes classification and Laplace smoothing are up

Hello class!

Notes from yesterday's recitation are up covering naive Bayes classification and Laplace smoothing (you'll need to know both for Problem Set 5). If you attended the 1-2pm recitation, I ran out of time and didn't talk about Laplace smoothing, so be sure to read up on Problems 2(b) and (c) in the recitation notes. Also, I switched up the notation a bit to have it be consistent with the problem set. The random variable for the class/label X (spam or ham) is now C, and the number of words in the dictionary k is now J. I'll make videos within the next few days.

Cheers,
George

Announced on 17 October 2012  12:18  p.m. by George H Chen

View archived announcements