Assignment 2
Probabilistic Models of
Human and Machine Intelligence
CSCI 5822
Spring 2018
Assigned Feb 1
Due Feb 8
Goals
The goal of part 1 of this assignment is to
give you experience in reading research articles involving
probabilistic approaches to machine learning. The goal of part 2
is to let you exercise your math chops dealing with multivariate
Gaussians
Part 1
Write brief commentary on the Weiss,
Simoncelli, & Adelson (2002) article that we discuss in
class on February 1. The commentary should be no more than one
page. The article is fairly dense -- both in understanding the
issues in human and machine vision as well as in the modeling
methodology -- but hopefully the overview of the paper I give in
class is sufficient to unpack the paper.
The commentary should start with a summary of what you think the
main or most interesting ideas are in the article. You do not
need to summarize the entire article. You can assume that
the reader of your commentary will have read the article, but
give your take on why others should be interested in it and what
research contribution it makes.
In addition to the summary, include one or more of the
following:
- Questions about the material for further discussion,
either clarification questions or points of disagreement
with the authors (``I don't see how such and such will work
as the author claims...'').
- Comments on how the assigned reading relates to other
material you've read, or, if you feel ambitious, you can
track down related work in the field and discuss how the
assigned article compares to this other work.
- A critique of the work, that might include: What
are the flaws in the ideas presented? What are the
limitations? Do the authors place their work in the
appropriate theoretical perspective? Do the authors
overstate their results?
- Your ideas for how the work could be extended or made more
interesting and more relevant.
Part 2
Consider the multivariate Gaussian
distribution
x ~ N(
μ, Σ),
where the vector
x has x
1 components
x1, x2, ...
xn. The definition
of this distribution is in my notes and in Definition 8.28 of
Barber (p. 172). Derive p(
x1 | x2,
... xn).
Hint: make use of Equation 8.4.19 in Barber (p. 174).
My title