Rensselaer Polytechnic Institute
Department of Computer Science


CSCI 6100/4100: Machine Learning

Fall 2008

ANNOUNCEMENTS

  • Nov 6: Final project handout is now posted. Due December 4.
  • Oct 23: Homework 2 is now posted. Due November 6.
  • Sep 26: Deadline for Homework 1 extended to Oct 2, start of class.
  • Sep 9: Homework 1 is now posted. Due September 29 October 2.

    • OVERVIEW

    This course is an introduction to many of the important themes in machine learning, including supervised and unsupervised learning, online learning, reinforcement learning, and learning in societies of agents. It will focus on statistical methods, how to evaluate success, and learning as a critical component of decision-making.

    Instructor: Sanmay Das
    Office: Lally 302
    Phone: x2782
    Office hours: Mondays from 3:30-4:30 PM and by appointment.

    SYLLABUS

    The syllabus is available here.

    LECTURES

    Lectures will be on Mondays and Thursdays from 2pm to 3:30pm in Low 3130.   
    Date Topics Notes
    Aug 25 Introduction to supervised, unsupervised, reinforcement, and "rational" (Bayesian) learning. Review of probability, statistics, and linear algebra. PDF notes
    Aug 28 Estimators. Maximum likelihood. PDF notes
    Sep 4 Bayesian estimators. Evaluating estimators: MSE and the bias/variance tradeoff. PDF notes
    Sep 8 Regression: Least squares and BLUE (see previous lecture notes). Confidence intervals for errors. PDF notes
    Sep 11 Naive Bayes and logistic regression. Tom Mitchell's book chapter
    Sep 15 Logistic regression continued. Evaluation and ROC curves. PDF notes
    Sep 18 Decision trees. ID3. PDF notes
    Sep 22 No class. Attend CS Day on machine learning!
    Sep 25 Pruning. Perceptrons. PDF notes
    Sep 29 Support vector machines. PDF notes
    Oct 2 Nearest-neighbor methods and ensemble classifiers (bagging and boosting). PDF notes
    Oct 6 Unsupervised learning. k-Means and Expectation Maximization. PDF notes
    Oct 9 Utility theory. PDF notes
    Oct 13/14 No class. SD at INFORMS
    Oct 16 In-class exam
    Oct 20 Temporal models: filtering and smoothing. PDF notes
    Oct 23 Temporal models contd: Viterbi algorithm, Kalman filters. Intro to Markov Decision Processes. MDP notes
    Oct 27 MDPs continued See previous notes
    Oct 30 Bandit problems PDF notes
    Nov 3 Reinforcement learning PDF notes
    Nov 6 Function approximation in RL. Optimal stopping. RL notes from previous lecture. Optimal stopping notes from next lecture.
    Nov 10 Optimal stopping contd. Intro to POMDPs. Optimal stopping reading: Chap 2 of Tom Ferguson's book
    Nov 13 More on POMDPs. POMDP reading: This paper by Kaelbling, Littman and Cassandra
    Nov 17 Online learning and experts. PDF notes
    Nov 20 Intro to game theory. PDF notes
    Nov 24 Market experiment.
    Dec 1 Learning in games.
    Dec 4 In-class exam
     


    TEXTBOOKS

    The course will not be based on any single book.   The following textbooks are all recommended as references:

    Many of the topics we cover may differ significantly in coverage from any of the texts mentioned above. Therefore it is important to come to lectures and take notes.

    ASSIGNMENTS

    Assignment 1: Due Sep 29 Oct 2
    Assignment 2: Due Nov 6
    Class project: Due Dec 4