6.867 Machine Learning -- Problem sets and quizzes

PS Due date Assignment text Data Software Solutions
1 Thursday Sep 19 mpg.dat
mpg.names
mpg_full.dat
trainNtest.m
train.m
predict.m
testLinear.m
degexpand.m
2 Thursday Oct 3 XYsmall.dat
XYtrain.dat
XYlarge.dat
XYtest.dat
You can download all the relevant files in an archive:
ps2.zip or
ps2.tar
Or donwload them one by one:
activelearn.m
fullexpand2.m
knn.m
knntest.m
logisticreg.m
logistictest.m
make2Dgrid.m
multigaussian.m
oracle.m
plotgaussians.m
plotlogreg.m
plotres.m
predictgaussians.m
3 Tuesday Oct 22 docdata.mat
detectors
NBest.m
NBclas.m
cv.m
selGreedy.m
4Thursday Oct 31 NoneNone
5Thursday November 14 MOGdata.dat
MOGsmall.dat
MOGlarge.dat
em_mix.m
bictest.m
6Wednesday November 27 data45.mat or
data45.zip or
data45.mat.gz
HMM code:
HMM.zip or
HMM.tar.gz

Spectral clustering:
spectral.m
mkdata.m


You can take a look at the last year's problems and exams at the last year's pages.

Collaboration Policy

Collaboration is encouraged while solving problems, however you must write your own solution to every problem. Please list your collaborators on every homework that you turn in. Copying will not be tolerated.

Exams

We will be holding a mid-term, on October 17th, and a final exam on December 5th. Both exams will be in class.

Project

Each student will be responsible for a project of their own selection (subject to approval). The projects are due on Wednesday, December 11th.

Grading

Your final grade will be determined from
  • Problem sets (30%)
  • Final project (30%)
  • Mid-term (15%)
  • Final (25%)
  • The percentages are approximate and may be changed at our discretion.