Questions and Comments for: K-Nearest Neighbors Density Estimation

A slecture by Dan Barrett

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.

This slecture is reviewed by Chuohao Tang

The slecture introduces density estimation and classification technique using K nearest neighbors method.

The slecture is well organized. The problem statement and solutions described are very clear. Using one example through out the slecture is fairly illustrative. Instead of just giving the formulas, the author gives some intuition thoughts and things needs to be paid attentions to behind these formulas to help readers better understand their meanings.

There are somethings need to be improved.

• The Bayes theorem equation is incorrectly typed. The righthand side should be divided by $p(x)$.
• In the equation where $V_k(x_0)$ is defined, the part $\sum_{l=0}^{N} \phi(\frac{x_l-x_0}{h}) = k$ I think it should not include $x_0$ itself. In other words, $l$ should start from $l=1.$
• The visualization of density estimation is not clear. The figure title doesn’t match the describing text. Only mentioned red and blue, not other colors.