Instructor: |
Prof. Alexandre V. Morozov Office hour: by request |
Prerequisites: Basic knowledge of linear algebra and probability theory. Homework and Exam: One homework per 2-3 weeks. There will be a final take-home project (72 hours, open book, open notes).
The grade is
determined according to the following formula: total score = 1/2(homework) + 1/2(final) Lecture 1 (01/20)
  pdf   online (Zoom link sent by email)
Lecture 2 (01/24)
  pdf   online (Zoom link sent by email)
Lecture 3 (01/27)
  pdf   online (Zoom link sent by email)
Lecture 4 (01/31)
  pdf
Lecture 5 (02/03)
  pdf
Lecture 6 (02/07)
  pdf
Lecture 7 (02/10)
  pdf
Lecture 8 (02/14)
  pdf
Lecture 9 (02/17)
  pdf
Lecture 10 (02/21)
  pdf
Lecture 11 (02/24)
  pdf
Lecture 12 (02/28)
  pdf
Lecture 13 (03/03 and 03/07)
  pdf
Lecture 14 (03/10 and 03/21)
  pdf
  Lecture 14 Supplement
  pdf
Lecture 15 (03/24)
  pdf
Lecture 16 (03/28)
  pdf
Lecture 17 (03/31)
  pdf
Lecture 18 (04/04)
  pdf
Lecture 19 (04/07)
  pdf
Lecture 20 (04/18)
  pdf
Lecture 21 (04/21)
  pdf
  [based on MacKay 42.1-42.5 (pp. 505-510)]
Lecture 22 (04/25 and 04/28)
  pdf
  [based on MacKay 42.7-42.8 (pp. 512-516) and 43.1 (pp. 522-525)]
  Amit et al PRL 1985
Lecture 23 (05/02)
  pdf
  [based on MacKay 43.1-43.2 (pp. 522-526)]
  Hinton & Salakhutdinov Science 2006
  SI
Homework 1 (due 02/17):
Problems
Homework 2 (due 03/07):
Problems
Homework 3 (due 03/31):
Problems
Homework 4 (due 04/21):
Problems
Homework 5 (due 05/02):
Problems
Note: NO lecture on Monday, Apr. 11
Note: NO lecture on Thursday, Apr. 14
Please
send any comments
about this page to morozov at physics.rutgers.edu
Textbooks:
Pattern Recognition and Machine Learning (Information Science and Statistics) by Christopher M. Bishop.
Information Theory, Inference and Learning Algorithms by David J. C. MacKay.
Reviews:
Introduction to Machine Learning for physicists by Pankaj Mehta et al.
Lecture Notes:
Homework:
Final Exam (due 05/09 by 5 pm):
Department of Physics
and
Astronomy Main Page