Revision as of 05:27, 9 May 2014 by Agheith (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Questions and Comments for: Introduction to Nonparametric (Local) Density Estimation


Please leave me comment below if you have any questions, if you notice any errors or if you would like to discuss a topic further.


Questions and Comments

[Review by Alex Gheith]

This slecture provides a quick introduction to the topic of local (nonparametric) probability density estimation. It provides an intuitive motivation as to why we need this technique, which I believe is a key point in any introduction. The presentation slides are very clear and well-written. The video and audio resolution are also very clear, which is NOT present in all slectures (including my own.)

One suggestion to bring up is that density estimation is a key element in pattern recognition, which is the main application of the course for which this slecture has been made for. However, both parametric and nonparametric density estimation methods are independent of this unique application, so I think a hint to this effect maybe useful to the general audience.

Another suggestion is that the plot in the slide titled "Local Density Estimation (1)" only shows ideal density plots, and it can make use of another laid-over plot of estimated density, i.e. a jagged version of the same densities, to show how the noise from the estimation can introduce an error in the decision boundaries.

In closing, I think this slecture succeeded in achieving the goal -- to provide an intuitive introduction of the topic to entry-level audience.

Alumni Liaison

has a message for current ECE438 students.

Sean Hu, ECE PhD 2009